[skimage] annotated tag v0.12.0 created (now a8cc4b5)

Andreas Tille tille at debian.org
Fri Dec 23 08:22:14 UTC 2016


This is an automated email from the git hooks/post-receive script.

tille pushed a change to annotated tag v0.12.0
in repository skimage.

        at  a8cc4b5   (tag)
   tagging  50fa219975df80fc1ae81486bb48ed1c0daf791f (commit)
 tagged by  emmanuelle
        on  Sat Mar 5 18:14:52 2016 +0100

- Log -----------------------------------------------------------------
0.12.0 release

Adam Feuer (11):
      pil_plugin can import and export from memory
      added tests for roundtrip export and import
      added a space character to trigger travis build
      fixed bad indent
      improved name of pil import/export functions
      reformat comments to be standard format
      added Adam Feuer (myself) to contributors file
      updated imread parameters comment to standard format
      removing trailing space (PEP8 formatting)
      explicitly checking for None as per code review feedback
      removed trailing blank line in comment

Adam Ginsburg (2):
      very slight change to warp_coords docstring, but I think slightly more illustrative of the shape of the array
      looks like I didn't save on the previous commit; these should be squashed with the previous commit message

Adam Wisniewski (4):
      support for saving to file-like object in imsave
      Merge pull request #1 from tonysyu/pr/613
      fix arraypad
      adding main to run tests

Alexandre Fioravante de Siqueira (2):
      Solving white space + improving code +PEP8
      Solving space convex_hull + fixing local_otsu

Alexey Buzmakov (1):
      Remove endpoints from theta

Alexey Umnov (3):
      Fixed ellipse symmetry and added some tests.
      Added float conversions for python2.x
      Updated CONTRIBUTORS.txt

Alexis Mignon (8):
      Modified 'denoise_bilateral':\n1) check that image max value is not zero to avoid segmentation fault.\n2) initialize color_lut after sigma_range has been checked.
      Deported variable initialization after the cdef block to allow verifications.
      Moved image conversion after the verification of the max_value
      Added non rotation invariant local binary patterns
      Corrected duplicate variable declarations, variable name change from 'var_index' to 'rot_index' and corrected minor bugs
      Finalized changes. Added reference for non rotation-invariant uniform patterns. Added test case. Modified comments to clarify the process, and changed the code to make it more consistent with explanations and actuall encoding
      Corrected name in reference [2]
      Changed link to the reference [2]

Almar (6):
      Create dummy WindowsError that is an actual Exception instead of None. The previous solution prevented installation on py3.3 and OSX 10.6.1.
      Add cube selem.
      Add ball selem. I verified in 2D that this usage of np.mgrid yields the exact same results as np.meshgrid (np.meshgrid is not available for 3D).
      Fix typo in docstring of disk().
      Adding octahedron selem (3D equivalent of diamond).
      Added test for ball and octahedton selems.

Almar Klein (18):
      Make mcp evolution algorithm a bit more readable.
      Improve speed of MPC by about 30%
      MCP: turn while loop into for loop to prevent infinite looping.
      MCP: add sampling attribute: deal with anisotropic data.
      MCP: refactored to use typed memoryviews
      update-mcp: Refactor MCP algorithm in prep for MCP subclasses
      update-mcp: Add MCP_Connect class
      update-mcp: Add MCP_Flexible subclass.
      update-mcp: Small fix.
      update-mcp: Change when examine_neighbors is called.
      update-mcp: add new MCP classes to skimage.graph namespace
      update-mcp: added tests for newly added features.
      update-mcp: improvements to docstrings.
      update-mcp: add tests for MCP_Flexible
      Update MCP: Put back the traceback() method.
      PEP8 stuff.
      Update docs for img_as_ubyte
      Update docs in dtype.py: mention scaling of values

Anders Boesen Lindbo Larsen (21):
      Add dense DAISY feature description.
      PEP8 style corrections.
      Better Sphinx documentation.
      Add circle perimeter to draw module.
      Add DAISY visualization + example plot.
      Handle all non-float images.
      Move set_color() to draw module.
      No need to normalize angular smoothing.
      Fixed submodule import.
      Add myself as contributor.
      Add dense DAISY feature description.
      PEP8 style corrections.
      Better Sphinx documentation.
      Add circle perimeter to draw module.
      Add DAISY visualization + example plot.
      Handle all non-float images.
      Move set_color() to draw module.
      No need to normalize angular smoothing.
      Fixed submodule import.
      Add myself as contributor.
      Merge branch 'master' of github.com:andersbll/scikit-image

Andreas Mueller (107):
      pep8
      removed dead link form canny help string
      check if input for canny is 2d array, test for error raise.
      somewhat sensible default values for canny edges
      pep8, removed unused import and variables
      new canny edge tutorial location
      rank filter asserts input is 2d
      rank filter has radius 2 by default. radius 1 fails ?!
      remove unused imports, avoid from x import *
      pep8
      adjusted default value test to new default values... what does that test do?
      in median filter, rename input data to "image"
      slightly faster image conversion by creating less temporaries.
      pep8 whitespace
      gracefully ignore absent plugins
      this depends on the plugin used
      removed debugging code
      [ENH] include matplotlib's imread
      try to load only plugins that have both imread and imshow
      add qt plugin again.
      COSMIT removed unused import
      DOC some sphinx and rst fixes
      DOC finally....
      DOC minor
      COSMIT pep8
      COSMIT don't let pep8 make your matrices ugly.
      COSMIT fix ugly line breaks.
      COSMIT minor pep8
      COSMIT no spaces around power ``**``. Fun: https://gist.github.com/1671995
      COSMIT minor stype improvements, whitespace in error messages
      COSMIT some manual pep8, removed unused imports, removed unused variables and fixed a bug in a ValueError statement.
      ENH slight cleanup, fixed *1000 bug, added test and rgb2lab convenience.
      MISC minor cleanup in doctests / examples
      Add segmentation setup.py for felsenzwalb algorithm
      ENH first draft of felzenszwalbs graph based image segmentation in Python
      ENH using union find from morphology module
      MISC remove debugging tracer, unnecessary variable.
      ENH naive pure python implementation of quickshift
      ENH fixed stupid bug in quickshift, example
      ENH start cythonizing quickshift, get rid of hstack.
      misc Uncomment Felzenszwalb as it is not messing with quickshift.
      ENH reasonable speed.
      Color example :)
      ENH CRAZY speedup
      Bigger example
      ENH dirty fix, works though. Starting profiling.
      ENH much faster.
      forgot a line :-/
      enh cythonizing some arrays
      ENH reasonable speed for felzenszwalbs's segmentation
      ENH felzenszwalb for color images
      ENH Polish examples.
      DOC more docs....
      misc remove profiling outputs from quickshift
      DOC document and export felzenszwalb_segmentation_grey, prettify plots for the web.
      ENH: make quickshift more tolerant to input type, just convert to float. Also keep track of random seed for reproducable tests.
      FIX build problem and cython problem resolved.
      enh: minor simplifications
      ENH Rename parameters in quickshift, add "ratio"
      Remove felzenszwalb_segmentation_gray again since it just complicates the interface.
      ENH tests for Felzenszwalbs segmentation, fixed off-by-one error
      First draft for numpy based km_segmentation
      FIX Tried to address @stefanv's comments on the PR.
      FIXed test to work with the fixed "hashing" of colors
      DOC updates CONTRIBUTORS.txt
      fix trying to make this implementation more like slic
      starting cython implementation of km_segmentation
      ENH speedup, means and image use pointers
      MISC some simplifications, minor speedup
      Documentation, example for km_segmentation
      Fix initialization in km_segmentation, prettier examples
      Rename km_segmentation to slic. They have a PAMI paper now so I guess we should use their name.
      Put RGB2Lab into slic as it seems to be essential.
      Some scaling issues to be closer to reference implementation
      quickshift: convert to lab in function, some comments in code.
      Fixed bug in SLIC smoothing
      Added smoothing option to quickshift
      merge segmentation examples
      Post-processing for felzenszwalbs algorithm.
      Pep8
      Fixup tests, add test for slic.
      MISC some typos in Example, titles set.
      pep8
      Minor fixes addressing @tonysyu's comments.
      FIX width/height trouble, add non-regression test
      ENH renamed "felzenszwalb_segmentation" to "felzenszwalb", remove debug output from slic
      Trying to avoid name collisions.
      DOC a bit nicer figure
      make figure smaller again.
      ENH minor speedups.
      ENH added some cdefs
      ENH more speeeeed
      ENH Felzenszwalbs segmentation somewhat faster
      Merge pull request #247 from ahojnnes/label-dtype-fix
      cosmit: changed the names of x and y to r and c. that was no fun, i tell you.
      ENH addressed (hopefully all) of Tony's and Stefan's comments.
      MISC move felzenszwalb_cy.pyx to _felzenszwalb_cy.pyx, don't use xrange when not necessary
      fixed data of tv_denoise result images with float datatype to be in default range [0.0:1.0]
      use existing functionality for fix
      convert image to float before performing tv_denoise operation now
      cleanup of tests hence all results of tv denoise operations are returned as float
      specified float array result in docstring
      MISC remove unused imports, some pep8 corrections.
      COSMIT some pep8
      FIX quickshift nosetest with different xyz color conversion
      FIX add missing min_size parameter, add regression test.
      slight performance improvement.

Andrew Mark (3):
      Fix missing peaks in hough_line_peaks.
      Local neighbourhood suppression on hspace_max.
      Sort peak candidates by intensity before starting supression.

Ankit Agrawal (228):
      Deprecating is_rgb()
      Reverting the changes
      Deprecating is_gray and replacing it with is_gray_2d
      Removing is_gray() and is_rgb()
      Removing redundant imports used in previous commits
      Deprecating is_rgb and is_gray
      Added a function for Gamma Correction
      Prevented info loss, implemented Log and Sigmoid Correction
      Separating the functions
      Separating correction methods into different functions
      Correcting code quality issues
      Added better tests and removed weak ones
      Added stronger tests
      Corrections and Improvements
      Minor improvements in docs, tests
      Error handling for images with negative values in Image Correction methods
      Adding a utility function to test non-negativity of an image
      Renaming rescale_intensity_* as adjust_*
      Replacing every occurence of substract* with subtract*
      Final stylistic changes in exposure.py
      Correcting typo in the reference section of adjust_sigmoid
      Correcting typo in the reference section of adjust_sigmoid.
      Adding author and name of the paper in reference section of adjust_sigmoid
      First quick implementation of BRIEF Feature descriptor
      Allowing arbitrary descriptor sizes in BRIEF
      Removing hard-coding in BRIEF code
      Adding custom sampling seed, Normal sampling mode
      Added docs and used numpy trickery to filter keypoints
      Changed dtype of hamming dist matrix and added docs
      Fixing indentation
      Merge branch 'ankit-brief' of https://github.com/ahojnnes/scikit-image into ahojnnes-ankit-brief
      Clearing docs and making feature.util
      Removing unused import and fixing indentation issue
      First implementation of Integer Up/Downsampling
      PEP8 corrections
      Removing code repetition
      Minor refactoring
      Downsampling of nD arrays
      Padding ndarray with zeros to support downsampling by any integer factor
      Added docs, tests for downsample() in skimage.transform._warps
      Making code compatible with Python 3
      Add _sum_blocks
      Cleaning up downsampling for integer factors
      Expanding the documentation in transform._warps
      Attempt to solve circular import issue
      Another attempt at solving circular import issue
      Correcting typos
      Added an example in skimage.feature._brief
      Fixing bug in _remove_border_keypoints
      Added more detailed example in docstrings
      Adding match_keypoints_brief
      Adding docs for match_keypoints_brief
      Making hamming_distance more generalized; improving docs
      Broadcasting in pairwise_hamming_distance; numpy optimization in match_keypoints_brief
      Added _brief_cy.pyx in bento.info
      Moving _brief_cy.pyx changes in bento.info to the features section
      Adding tests for BRIEF and pairwise_hamming_distance
      Fixing bugs and making code compatible with Python 3
      Final changes
      Replacing np.all by equivalent np.any
      Stylistic changes
      Added if __name__ == __main__ in new test files
      Adding Octagon structural element
      Removing circular import
      Correcting grammatical typo
      Making the doc more explicit
      Added star structural element
      Minor doc changes
      Test for Octagon
      Improving the implementation of star
      Adding namespace
      Adding test for star
      Adding tests for smallest selem
      Added partial implementation of censure for mode=DoB
      Line Suppression of the response
      Correcting import for a private function
      Adding TODO's
      Added mode Octagon using np.convolve
      Correcting typo
      For loop for mode=DoB in Cython
      Commenting out _filter_using_convolve
      Removing time() statements and correcting types in censure_cy
      Removing hard-coding in censure_keypoints()
      Returning keypoints with the scale info
      Correcting the expression of response in censure_keypoints()
      Correcting the definition of Ratio of Principal Curvatures
      Adding _slanted_integral_image. Removed _oct and _filter_using_convolve
      Removing unused stuff
      Cleaning up the non-Cython version for mode=Octagon
      Cythonizing the for loops
      Adding the docs for censure_keypoints
      Trying to resolve Cython TypeErrors
      Trying to Debug the SegFault
      Trying to Debug SegFault : 2
      Trying to debug the Segfault-3 : Making the arrays C-contiguous
      Fixing many bugs; Working Censure for mode=Octagon
      Adding doc for the different modes of _slanted_integral_inage_modes
      Correcting the docs and making them explicit
      Commenting the sigma passed to _suppress_line
      Removing unnecessary data-type conversion
      Removing unnecessary array pre-allocation
      Changing no_of_scales to n_scales
      Renaming *_wt to *_weight; correcting the object type of n in _censure_dob_loop
      Documenting the code; Removing ascontiguousarray statements
      Correcting indentation
      Using convolve in _get_filtered_image for mode=Octagon
      Removing slanted_integral_image Cython functions for mode=Octagon
      Reverting back to non-debugging mode in Cython
      Remove all the slanted integral image functions and the vivid graphics from censure.py
      Wrapping the long lines
      Added a NOTE explaining the preference of convolve over slanted integral image
      Adding the STAR mode
      Making the weight calculation statements more readable
      PEP8 corrections
      Replacing threshold by nms_threshold
      Returning the keypoints and scales separately
      Changing the variable names of constant objects
      Correcting a bug in Line Suppression
      Putting constant variables outside the definition
      Making the docs more explicit
      Correcting the imports
      Assigning boolean to the feature mask in line suppression
      Actively filtering border keypoints on all scales
      Adding tests for all the modes in censure
      Debugging Travis error
      Reverting back to last but one commit
      Making the code compatible with Python 3
      Actively filtering border keypoints for all the scales part 2
      Removing print statements
      Returning mask in _remove_border_keypoints
      Filtering out border keypoints using masking
      Replacing int32 by intp
      Changing _remove_border_keypoints to _mask_border_keypoints
      Lowercasing variables; removing unused import; replacing _remove by _mask
      Replacing censure_keypoints by keypoints_censure, n_scales by max_scale
      Documenting the code
      Including min_scale as another input parameter
      Adding remaining tests
      Adding gallery example for plotting censure keypoints
      Adding a short description of the algorithm
      PEP8 corrections and stylistic changes
      Adding Octagon shapes for higher scales
      Importing star and octagon from morphplogy.selem
      Correcting selem.star for Python 3
      Reverting back to changes made by Johannes
      Final changes in censure example
      Documenting the choice of filter sizes
      Correcting bug in SimialarityTransform
      Quick initial implementation of ORB descriptor in Python
      Quick Python implementation of FAST corner detector
      Implemented FAST corner detector
      Adding docstrings and commenting the code
      Adding high speed test for n>=12
      Making various minor optimizations
      Transferring all the FAST code to corner.py and corner_cy.pyx
      Using inline function in corner_cy._corner_fast
      Deleting FAST files
      Converting image to float
      Adding additional parameter for setting shape for Non-max suppression
      Wrapping lines; returning corner_response at a pixel
      Returning the corner response
      More optimizations
      Quick initial implementation of ORB descriptor in Python
      Quick Python implementation of FAST corner detector
      Implemented FAST corner detector
      Adding docstrings and commenting the code
      Adding high speed test for n>=12
      Making various minor optimizations
      Transferring all the FAST code to corner.py and corner_cy.pyx
      Using inline function in corner_cy._corner_fast
      Deleting FAST files
      Converting image to float
      Adding additional parameter for setting shape for Non-max suppression
      Wrapping lines; returning corner_response at a pixel
      Returning the corner response
      More optimizations
      More clear docs; Wiki reference; removing rebase leftover
      Implemented corner_fast_orientation function
      Using Py_ssize_t and np.intp for cfast_corners
      Adding tests for corner_fast and corner_fast_orientation
      Speeding up corner_fast_orientation
      Speeding up corner_fast_implementation : 2
      Adding an example; removing redundant stuff and more
      Making changes in test_corner
      Making the example more explicit
      Inserting corners in the corner_orientaions example
      Implementing keypoints_orb without using Harris response
      Returning best_keypoints
      Docstrings; Stacking lists; removing redundancy in corner_orientations
      Adding explicit test for corner_orientations
      Implementing descriptor_orb; for loop in Cython
      Returning keypoints; Adding docstrings
      Minor doc change
      Better naming of arguments
      Writing better Cython code
      Stacking lists in descriptos_orb
      Adding example in descriptor_orb
      Converting descriptors dtype to bool
      Minor optimizations; Renaming variables; Docstring corrections
      Adding square examples in keypoints_orb, descriptor_orb
      Introducing the _prepare_grayscale_input_2D function in feature.util
      Making pos0 and pos1 as global variables
      Fixing the import issue
      Better wrapping of lines; making ofast_mask a global constant
      Improved match_binary_descriptors function
      ORB matching example
      Minor code changes; Explicit docs
      Changing values of default parameters and OFAST_MASK
      Correcting the orientation bug
      Making matching graphics more clear
      Documenting the example code
      Minor doc change
      Minor doc changes
      Incorporating mat_binary_descriptors changes in test_brief
      tests for match_binary_descriptors; many corrections and changes
      Cleaning util.py
      Adding tests for orb; fixing some bugs
      Fix missing import
      Absolute path fix
      Introducing recarray for storing keypoint variables
      Made recarray changes in docstrings and tests
      Extending recarray changes to BRIEF
      Incorporating recarray changes to match.py
      Made recarray changes in the brief docstring
      Recarray changes made in example
      Correcting the strides of response in _filter_image
      Fix regionprops doc typo
      Adding test to check Censure works on any 2D shape

Antony Lee (4):
      Use relative imports to plugins.
      Saner defaults for peak_local_max, untested.
      Change default min_distance; improve docstring.
      WIP: fix impl. of peak_local_max, update tests.

Arnaud De Bruecker (2):
      Add documantation to `denoise_bilateral` for `sigma_range`
      Correct documentation to `denoise_bilateral` for `sigma_range`

Blake Griffith (19):
      Add process_chunks function
      load process chunks into namespace
      Tests for process chunks function.
      Add latest dask to .travis.yml
      Add latest dask to appveyor
      pep8
      Smarter chunk selection.
      Move the function's args and kws to the end of the signature
      process -> apply, more docs
      Use my dask branch instead of master
      Rename apply and apply_chunks to apply_parallel
      Install dask[array] from pypi for travis and appveyor
      Use absolute import
      Add dask to DEPENDS.txt
      Add to release notes.
      Defer import errors of dask array until runtime.
      Move dask intall to travis_befor_install.sh
      Remove dask install from travis_script.sh
      Correct link in DEPENDS.txt

Brian Holt (17):
      Initial implementation of Histograms of Oriented Gradients
      pep8 + addressing some of Stefan's comments
      replaced gradient operation with np.diff
      replaced arctan with arctan2
      merged upstream master
      use sat_sum for performance improvement
      Merge pull request #1 from amueller/hog
      Initial implementation of Histograms of Oriented Gradients
      pep8 + addressing some of Stefan's comments
      replaced gradient operation with np.diff
      replaced arctan with arctan2
      use sat_sum for performance improvement
      added HOG example
      pep8
      added HOG example
      pep8
      simplified hog code + extra unit test

Charles (2):
      DOC : add ref to nlmeans
      DOC : add ref to nlmeans in the API

Charles Deledalle (1):
      DOC : change sneaky unicode in the endash

Charles Harris (1):
      BUG: Fix slicing error revealed by numpy 1.9.0.

Chintak Sheth (34):
      Clean, and implemented using horse.jpeg
      Added horse.png to skimage/data and added a horse() in the _init_.py in data subpackage
      End newline added and removed the .jpeg version
      URL formatting
      Minor changes
      Trailing _
      URL
      URL
      .in to .com
      Added convex_hull_object function
      Trailing whitespace
      Added the note in docstring as suggested by Josh
      Update convex_hull.py
      Note and relative import
      Added unit test for convex_hull_object function in test_convex_hull.py
      Unit tests added
      Resolve fast forward
      Remove trailing whitespace
      Remove trailing whitespace
      Using labeled_im == i
      Blank at the end of docstring
      Add convex_hull_object in __init__.py
      Minor changes
      Ability to choose the type of connectivity and ValueError
      Test for the two neighborhood values
      Note rewritten
      Endline
      Added test for checking ValueError
      Note re-worded
      Whitespaces and note
      Space
      Add convex_hull_object to __all__
      new line
      Docstring for _dilate and _erode were interchanged

Chris Beaumont (1):
      vectorized transform.integral.integrate

Christian Sachs (4):
      added faster, Cython based skeletonize function
      fixed docstrings, renamed variables for clarity, removed old skeletonize function body
      reordered code according to PEP8, modified docstring
      fixed intra docstring reference links

Christoph Deil (1):
      Some improvements to new deconvolution code

Christoph Gohlke (14):
      Do not install the doc package
      Use log(2) instead of log10(2)
      Use log(2) instead of log10(2)
      Fix `ValueError: Buffer dtype mismatch` on win-amd64
      Fix `ValueError: Buffer dtype mismatch` on win-amd64
      Fix `ValueError: Buffer dtype mismatch` on win-amd64
      Fix `ValueError: Buffer dtype mismatch` on win-amd64
      Use np.intp type for indices
      Fix MSVC error C2036: 'void *' : unknown size
      Fix skimage.io.imread returns incorrect dimensions
      TST skimage.io.imread returns incorrect dimensions
      Fix failing test
      Fix failing test
      DOC: add WinPython distribution; correct spelling

Christos Psaltis (2):
      Fix resize argument handling in rotate.
      Add test for resize option in rotate

Colin Lea (1):
      Swapped the direection that the histograms point in the hog visualisation

Connelly Barnes (2):
      fix scikit-image bug #1680 (grayscale png round trip fails) and add unit test
      fix permissions issue in test_pil.py function test_png_round_trip()

Daichi SUZUO (1):
      Fix a broken link to daisy code

Damian Eads (7):
      Finally, I've checked in my additions (hopefully) the right way.
      Fixed docstrings. Lines now only span <= 80 chars.
      Fixed dtype docstrings.
      Based on scikits/image/opencv/setup.py
      Added subpackage 'morphology' in setup.py so it traverses there.
      Regression tests written for morphology
      Regression tests for morphology

Dan Farmer (11):
      Merge Cell Profiler Canny edge detector
      Add init entry so importing works right
      Format for pep8 compliance, improve documentation, and make mask optional.
      Remove smooth.py for now
      Update docs and convert convolution to ndimage sobel function
      Converted logical ops to bitwise ops and made cosmetic changes wrt pep8 compliance
      Fix docstrings and one more pep8 issue
      Modified load_library for windll
      Fix indentation error that broke linux support
      Fix more indentation errors
      Remove debugging text

Daniel Wennberg (5):
      Fix tests for #1257
      Fix doctests for #1257
      Change data types to conform to style guide
      Fix PEP8
      Add note about blob detection api change

David Cournapeau (3):
      FEAT: new shot at bento build.
      FEAT: update to bento master - simplify waf declaration.
      REF: tweak bento build.

David Koeller (2):
      Fix notebook display bug for novice module
      Remove Picture.scale and Picture._rescale from novice module

David PS (2):
      fixed another typo
      Titles of the plots for the wathershed example not seeing for the margins of the plot.

Davin Potts (1):
      Modified test_rank.test_compare_autolevels and test_rank.test_compare_8bit_vs_16bit to correct for assumption that data.camera() would always be loaded as ubyte (or ints at least) when using matplotlib as backend (which caused data.camera() to load as float32).

Dharhas Pothina (5):
      added xyz2lab, lab2xyz not quite working
      fixed array being modified in place. lab2xyz working
      tests added for xyz2lab and lab2xyz
      minor cleanup of comments & move ref_white to global lab_ref_white
      Merge pull request #2 from amueller/dharhas-lab-color-space

Dhruv Jawali (1):
      removed reduntant continue, added Cython flags and fixed negative indexing

Egor (2):
      FIX/TST: Fixed bug with defected bounding frame, added testcase
      FIX: Fixed test for biharmonic in accordance to new kernel for edges

Egor Panfilov (14):
      DOC: plot_edge_modes improvement, closes #1816
      ENH: Inpainting with biharmonic equation
      DOC: Added gallery example for inpainting
      TST: Attempt to fix Travis bug with old packages
      DOC: Moved references to common docstring
      ENH: Added support for multichannel arrays, fixed float scale to [0;1]
      DOC: Fixed docstring, moved to assert_nD
      DOC: Moved gallery example to category
      ENH: Massively reworked code to enable nD-support
      ENH: Moved nested functions, refactor code
      FIX: Returned handle for empty input arrays
      TST, DOC: Added test for 3D spatial data, cleaned up gallery example
      DOC: Added credits for #1804
      Merge pull request #1895 from OrkoHunter/denoise_bilateral

Emmanuelle Gouillart (83):
      Give explicitly the git clone command in the installation notes.
      New module for total variation denoising, for 2D and 3D arrays.
      Moved the tests of tv_denoise to filter/tests
      documented the input dtype in tv_denoise
      Modified README to point to the new github repo. Mentioned in install doc
      Modified README to point to the new github repo. Mentioned in install doc
      Known errors during build for Windows.
      Fixed broken plot_directive.py so to build the doc.
      Merge branch 'master' of https://github.com/scikits-image/scikits.image
      Name of new github repository in setup.py of submodules
      Example generation from scripts, taken from scikit learn
      A few fixes: give credit to scikit-learn guys, fix examples...
      Change default angles in iradon (as many angles as projections)
      More tests on radon and iradon
      Merge branch 'stefan-radon'
      Renamed fast_watershed to watershed.
      Docstring for rank_order
      BUG: division by zero for small arrays in iradon
      Modifications to random walker segmentation algorithm:
      Better handling of labeled pixels in random walker segmentation when we
      [BUG] fixing import problem
      ENH: better handling of labels that need to be reordered
      PEP8: indentation in random_walker_segmentation
      Fixed two tests that were failing because of draw imports
      acronym --> full name of algorithm in example
      DOC: example in slic segmentation docstring
      PEP 8: _quickshift module
      Example in route_through_array
      DOC: docstring of route_through_array
      Merge pull request #248 from ahojnnes/clear-border
      Merge pull request #281 from mkcor/harris_threshold
      Merge pull request #284 from ahojnnes/regionprops-coords
      Merge pull request #285 from ahojnnes/label-example
      Merge pull request #282 from JDWarner/multispectral_random_walker
      Merge pull request #278 from wuan/tv_denoise_fix
      DOC: removed unused import in docstring
      [BUG] Corrected a bug in the random walker that appeared
      median filter outlier
      histogram example
      Find contours example
      Blank lines in reconstruction docstring for nicer display
      Mentioned the Neurodebian repo in the install doc
      DOC: Comment on the behavior of ``histogram`` for color images.
      Merge pull request #584 from josteinbf/iradon-algebraic
      Merge pull request #635 from josteinbf/radon-cleanup
      Some additions to the user guide: getting started section +
      Minor modifications following review by Johannes
      Merge pull request #672 from jni/chull-fix
      Merge pull request #710 from jni/pyamg-depends
      Added a wrapper around ndimage's Gaussian filter
      DOC: corrected description of multichannel parameter
      [BUG] iterable sigma and multichannel
      Gaussian filter function: changed the default value of multichannel
      Merge pull request #469 from JDWarner/add_marching_cubes
      Merge pull request #1119 from ikajic/filter-docu
      2-D and 3-D implementation of NL-means denoising
      Some improvements of non-local means denoising:
      Modified indentation in places (PEP8)
      Merge pull request #1244 from ahojnnes/rank-fix
      Merge pull request #1306 from stefanv/py3_apigen_fix
      Merge pull request #1675 from dpshelio/titles
      Merge pull request #1660 from jwiggins/issue-1586
      Merge pull request #1631 from stefanv/rename_denoise_nl_means
      Merge pull request #1663 from juliusbierk/hough-circle-example
      Merge pull request #1659 from rougier/gallery-cleaning
      Merge pull request #1690 from sciunto/favicon
      Merge pull request #1704 from MartinSavc/example_subplots_share_axes
      Merge pull request #1783 from ivoflipse/fix_hog
      Merge pull request #1826 from jni/rag-generic
      Merge pull request #1782 from pratapvardhan/clncython
      Merge pull request #1845 from grlee77/tv_weight_fix
      Merge pull request #1777 from stevendbrown/rp_intensity_check
      Merge pull request #1810 from alexandrejaguar/master
      Merge pull request #1842 from jni/1840
      Merge pull request #1846 from arokem/hull-docs
      Merge pull request #1806 from JDWarner/sphinx_formatting
      Merge pull request #1850 from jwiggins/fix/potential-bilateral-leak
      Merge pull request #1705 from deledalle/ref_nlmeans
      Merge pull request #1747 from JGoutin/OptimizeLevel2Compilation-patch
      Merge pull request #1883 from soupault/credits_pr1804
      Merge pull request #1920 from sciunto/doc_typo
      Merge pull request #1177 from stefanv/polygon_perimiter
      Merge pull request #1924 from ev-br/tempita

Endolith (1):
      DOC: Change some `jet` colormaps to `gray`

Eric (3):
      Blob detection supports for non-integer sigmas
      Changed local_maxima to float32
      Fixed typo

Eric Dill (1):
      TST: Barn-door testing of SaveButtons clicks

Evgeni Burovski (42):
      ENH: handle tempita templated Cython source files, *.pyx.in
      BLD: use default encoding
      MAINT: address review comments
      ENH: start adding a python version of Skeletonize3D code
      MAINT: move new skeletonize3d files to more appropriate places
      MAINT: skel3d: move computations to cython
      MAINT: skel3d: add static typing to obvious places
      MAINT: skel3d: only allocate and type neighborhood once; switch off
      MAINT: skel3d: statically type index_octants
      MAINT: skel3d: statically type Euler LUT
      TST: skel3d: add a Bat Cochlea sample from FIJI collections
      MAINT: avoid doing np.sum in the hot loop
      MAINT: avoid incrurring np.r_ overhead in is_simple_point
      MAINT: statically type `img` in toplevel _compute_thin_image
      MAINT: skel3d: simplify dealing with 2D images
      DOC: skel3d: add basic docstrings
      MAINT: skel3d: don't print stuff
      MAINT: stop slicing memoryviews in is_simple_point.
      STY: move main functions to the top of the file
      MAINT: remove decorators which have no effect anyway
      MAINT: skel3d: prune dead code (eg, is_surface_point)
      STY: skel3d: clarify the status of module-level LUTs.
      MAINT: skel3d: address review comments
      STY: reformat the filling-in of the Euler LUT
      MAINT: skel3d: rename _skel.pyx to _skel.pyx.in
      STY: generate `octree_labeling` routine via tempita at build time
      MAINT: skel3d: minor sty/perf fixes
      ENH: skel3d: unroll and inline `index_octants` into is_euler_invariant
      BLD: add the extension to bento.info, fix names
      MAINT: skel3d: address review comments
      BLD: skel3d: eradicate stray unicode symbols from source files
      BLD: bump the cython version to 0.23.4 on Travis
      TST: skel3d: add more tests
      DOC: spell out the names of [Lee94], add the Notes section
      MAINT: skel3d: do not copy more than necessary
      TST: inline a two-hole image
      TST: skel3d: remove 2D test images
      TST: skel3d: redo 3D test images
      MAINT: address review comments
      MAINT: address review comments
      STY: more tweaks in response to review comments
      BUG: fix the conversion from .pyx.in to .pyx files

Fedor Baart (1):
      setup

Ferdinand Deger (3):
      Added isotropic TV denoising + comment
      minor changes in isortopic tv denoising
      eleminate empty line

François Boulogne (232):
      add andres method for circle_perimeter
      fix variable
      fix docstring + PEP8
      add references
      new method: ellipse_perimeter
      add myself to contributors
      remove print statement
      remove brackets
      fix for master
      fix docstring
      add test point position circle-ellipse
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      Docstring circle + ellipse
      hough transform for circles
      add unittest for hough circle
      update contrib
      add blank line
      various minor fixes
      declare cython variables
      use a single array
      variable in cdef
      bugfix dim + unittest
      add example for cicular hough transform
      Docstring + J.S. modifications
      hough transform for circles
      use draw instead of mpl patches
      remove import mpl patches
      fix comment
      DOC: minor fix ellipse_perim.
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      DOC: add docstring for deprecation
      Fix mistake in example
      Merge branch 'docstring' of github.com:sciunto/scikit-image
      add note for distance/angle convention. Thanks to Emre Safak.
      replace hough() by hough_line() in doc
      remove docstring
      be consistant with python implementation
      improve unittests
      Fix and improve example
      Fix docstring
      minor doc improv.
      doc, minor improv.
      big clean-up!
      fix names
      fix signature
      cosmectics (flake8)
      fix deprecated variable
      Tony's comment on output format
      add option full_output to hough_circ
      fix travis complaint
      readability
      implement rotated ellipses
      add contribution
      remove old line
      minor fixes
      change bezier_curve definition for consistency
      split tests
      docstring
      rename function bezier_curve to bezier_segment
      add unittest
      pep8
      fix int + pep
      use bint
      oops
      add optional + default in pyx docstring
      MIN: use print() instead of print
      MIN: fix unused import
      fix blank print for python2
      cast range to list for python3
      minor contrib.
      MIN: remove unnecessary brackets
      import configparser for python3
      fix path import viewers
      fix import urllib2 for  python3
      workaround for itervalues
      fix iteritems for python3 compat
      use items/values for python2/3 compat
      Hough transform for ellipses
      use memoryview
      s/int/long memory view
      delete duplicated import
      fix imports
      remove commented lines
      fix import
      delete comment
      fix import testing
      add spaces for travis (pep8)
      fix import testing
      pep8
      fix missing space
      PEP8 + import
      activate travis for a fork
      erode & dilate -> private functions
      DOC: add doctrings
      MIN: fix pep8
      use previous import
      add coffee picture from flickr
      add example HT for ellipse
      add extra comments
      fix title
      rename files
      fix doctest
      PEP8
      add original picture
      simplify modification image + comment
      fix Johannes' comments
      use shorter URL
      typo
      revert, accum is a tuple of lists
      detailed procedure
      minor modifications
      DOC: changes for Marianne's coffee cup
      DOC: protect last character in URL
      Review the procedure
      DOC: this step seems to be useless
      DOC: add titles to plots
      import Fedor's contribution
      TEST: add unittest and check with travis
      DOC: update return according to wu's method
      TEST: fix returned tuple
      TEST: fix unittest
      DOC: fix refs
      MAINT: put AA method in draw.*_aa
      DOC: remove wu in methods
      Implement line_aa
      Anti-aliasing
      MAINT: bezier_segment is private
      MAINT: val in [0,1]
      DOC: split non-AA/AA
      FIX: division for value
      PEP8
      PEP8
      ADD: bezier_curve
      DOC: add bezier_curve
      MINOR: doctrings + various improvements
      FIX: broken test (thanks unittest!)
      MINOR: some comments
      PEP8
      PEP8: comparison
      MAINT: unused import
      MAINT: matplotlib is in requierements.txt
      DOC: no import numpy in doc
      DOC: add missing import
      PEP8: fix indentation
      PEP8
      DOC: fix indent
      TEST: fix mistake
      FIX: fix angle convention
      MAINT change HT return API
      ENH: use heapq to select the best match
      TEST: fix precision
      TEST: fix bad subs
      TEST: revert precision
      MINOR: fix non ascii char
      add missing import heapq
      FIX: handle correctly main axis def
      fix example according to new API
      remove heapq
      MAINT: np.pi -> M_PI
      MINOR: add missing type
      FIX: import M_PI
      PEP8
      PEP8
      PEP8 another one
      DOC: fix syntax error
      DOCTEST: fix
      DOC: remove deprecated paramter
      DOC: viewer: some corrections
      MAINT: populate __init__.py
      MAINT: remove 2 weird lines
      MAINT: remove blank lines
      TEST: add test on step
      DOC: specify range for step
      TEST: add unittest about ValueError
      MAINT: use asset_allclose instead of round
      remove extra )
      remove '' around ValueError
      MIN: typo
      DOC: typo
      PEP8
      FIX bug #832 + testcase
      MAINT: python3 compat.
      MAINT: use region.bbox
      MAINT: use region.area
      DOC: simplify the code (Thx Stefan)
      update info about PIL
      forgot a aword
      add step 4: review
      Stefan's comments
      Add Juan's comments
      Merge pull request #3 from stefanv/contributing
      fix typo
      add user guide parallelization
      Comments of the PR
      Stephan's comments
      simplify notations
      DOC: introduce variable num_peaks for reusability
      add shape option to draw functions
      MAINT: mutualize function
      DOC: enhance docstring
      DOC: fix mistake optional flag
      DOC: add link to main website + refresh
      fix phrasing
      DOC: move website and ML to the top
      DOC remplace iff by if and only if
      DOC: fix typos
      DOC: fix typo
      PEP8
      DOC: add parallelization to tutorials index
      add link joblib website
      attempt to fix #1567
      [DOC] add favicon
      [DOC] use favicon generated from skimage-branding
      DOC: activate favicon in config
      DOC: favicon path relative to source/
      MAINT: remove check for TRAVIS_PULL_REQUEST (never false)
      DOC: fix env variable GH_TOCKEN
      DOC fix broken url
      MAINT: use with to close the file
      MAINT: use with to close the file
      DOC: test piwik with piwik.sciunto.org
      DOC: test piwik with piwik.sciunto.org
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      Merge branch 'master' of git://github.com/scikit-image/scikit-image
      DOC: fix for typo
      use axis off
      MAINT: fix test name
      DOC: make image smaller for better visu
      MAINT: fixes #1652

François Orieux (44):
      Add deconvolution module to skimage.
      Rename test to tests
      Remove anfft and fftw dependency
      Use assert_allclose for testing.
      Remove __author__ line and add in CONTRIBUTORS.txt
      Change copyright to scikit-image team
      Add a small note explaining the wiener filter.
      Add pdf paper
      Inline crandn function.
      Whitespace...
      Fix the bad results of unsupervised deconvolution.
      Typo fix.
      Add test for Richardson Lucy.
      Add the reference image for Richardson Lucy test.
      Fix example.
      Fix relative uft import for travis-ci pass
      You shall not pass !
      Use skimage.img_as_float for travis-ci test
      Diminish rtol to 1e-5 for travis-ci
      rtol to 1e-3 for travis-ci.
      Rename deconvolution to restoration
      Fix test_restoration.py rename.
      Fix skimage/setup.py
      Typo and doc fixs.
      Fix doctest
      Fix circshift. Rename wiener.py to deconvolution (no API change).
      Small cleanup, keywords changes.
      PR review.
      Blank line remove
      Docstring convention about "Input image".
      Remove import numpy from docstring doctest
      PR fixs.
      Fix iscomplex to iscomplexobj. Replace circshift with numpy roll.
      Add doctest to uft.
      Change example scaling. Fix ref in doc.
      Better handle of non real case. Some docstring fix.
      Better docstring. Better test coverage.
      Add tests, coverage improvement.
      Fix travis test
      Fix python3 test and real values image.
      Docstring fix. Clip parameter addition.
      Remove own licence for skimage licence
      Update test file because of the new clip param default to True
      Remove credit field (and others). Pep8 small fix.

Gael Varoquaux (1):
      BUX: in _slic.pyx slice can has steps=None

Gael varoquaux (1):
      MISC: Clean ups in TV

Gary Ruben (1):
      Added forward and inverse 2-D Finite Radon Transform

Geoffrey French (13):
      Fixes a bug in _update_doc in skimage/io/__init__.py that will attempt to compute max of empty list if no plugins are found.
      Modified rank filters package so that the _core function in core.pyx outputs to a 3D image, permitting the generation of images with arbitrary size feature vector pixels. Implemented windowed_histogram that generates a windowed histogram of an image.
      Fixed some test failures and added a test for windowed_histogram.
      Fixed some issues with the new windowed_histogram function in filter.rank.
      Fixed Python 3 error in plot_windowed_histogram caused by use of incorrect division operator. Also tweaked the similarity computation a little.
      Exchanged array output parameter for pointer and Py_ssize_t in kernel functions for slight speedup.
      Improved plot_windowed_histogram doctoring, removed unnecessary imports and removed an internal testing line.
      Fixed comment in plot_windowed_histogram.
      Docstring and comment improvements and fixes in plot_windowed_histogram.
      Small fix to windowed_histogram doctoring.
      Fix to skimage.filter.rank.windowed_histogram docstring.
      PEP8 compliance and doc formatting fixes.
      Fixed mistake in generic.py

Gregor Thalhammer (18):
      unwrap: initial commit
      unwrap: compiles, test passes
      unwrap: cleanups
      unwrap: compiles ok
      unwrap: moved global variables (wrap_around) to params struct
      unwrap: implemented wrap_around arguments
      unwrap: changed mask values to numpy.ma convention
      unwrap: variable renaming
      unwrap: accept (and return if given) an masked array as input argument
      unwrap: added original 3D unwrapped code
      unwrap: moved pure python code from unwrap2D.pyx to unwrap.py, allocate output array in pure python
      unwrap: removed use of global variables
      unwrap: added 3D unwrapping (untested)
      unwrap: unified 2d-3d interface, 2d tested ok,
      unwrap: 3d tested ok
      unwrap: made it compile with MSVC (but crashes), still ok with gcc
      unwrap: added numpy test cases
      unwrap: add unwrap.py to python modules in setup.py

Gregory R. Lee (66):
      add CheckBox widget to viewer.  fix ComboBox widget docstring
      added tv_denoise.py viewer example using the CheckBox
      fix image scaling in tv_denoise.py example
      code formatting fixes
      PEP8 formatting
      remove outdated comment line
      update structural similarity to nD implementation with gaussian weighting to match other reference implementations
      add multichannel support to structural_similarity
      structural_similarity: add option to return full ssim image in addition to the mean
      structural_similarity: add image_content_weighting option
      sample vs. population covariance difference addressed to more accurately match reference implementations
      allow remaining hardcoded constants (sigma, K1, K2) to be modified via kwargs
      add additional tests
      replace _discard_edges with skimage.util.arraypad.crop
      revise argument handling code for multichannel case
      docstring fixes
      remove multichannel magic and default to False.  fix bugs in tests introduced during rebase
      use scipy.ndimage.gaussian_filter instead of a custom implementation
      remove image_similarity; set 11-tap Gaussian via truncate=3.5; add links to references
      remove truncate argument from gaussian_filter for scipy<0.14 compatibility
      BUG: fix bugs in coord_map for modes wrap and reflect
      TST: add tests for coord_map function
      DOC: add visual example of edge modes
      BUG: fix bento.info
      DOC: add visual example of edge modes
      BUG: bugfix for resize: upgrade image to 3d if necessary
      ENH: add mode 'mirror' to interpolation.pxd and dependent functions
      TST: add test for resizing 2D image with a 3D output shape
      DOC: update the edge mode example to include mirror
      MAINT: PEP8 fixes and _interpolation_test.pyx -> interpolation.pyx rename
      BUG: fix bento.info again
      BUG: fix import in plot_edge_modes.py
      MAINT: convert additional numpy.pad mode names to their dask.array equivalents
      MAINT: All modes in _shared.interpolation.pxd were changed to be consistent with numpy.pad naming conventions.  Specifically 'nearest' was changed to 'edge' and 'mirror' was changed to 'reflect'.  All functions with a mode argument that rely on these functions had their inputs changed accordingly.  For now there is a deprecation warning if the user supplies mode 'nearest'.  Mode 'mirror' never appeared in an official release of skimage and so has no corresponding deprecation warning.
      BUG: update the edge modes in the example to match the current naming convention
      MAINT: minor spacing issues and typos fixed
      BUG: fix one additional location where ndimage mode conversion was needed
      DOC: update TODO.txt to mention eventual removal of _mode_deprecations
      MAINT: remove unused import
      Change default weight for denoise_tv_chambolle to a value that is reasonable for float images scaled to the [0, 1] range
      Update tests to also use reasonable weights for TV denoising
      ENH: n-dimensional refactor of TV denoising.
      fix numpy 1.6 compatibility
      update tests for denoise_tv_chambolle
      remove try/except in _denoise_chambolle_nd.  just use the numpy 1.6 compatible case
      use Fortran order for g and p in _denoise_chambolle_nd to avoid a performance regression
      Switch _denoise_tv_chambolle to use C-ordering
      minor bugfix so that weight gives a consistent amount of smoothing regardless of the number of dimensions.
      update default weight to reflect the bugfix from hardcoded 0.5 to 1/(2*ndim)
      MAINT: explicitly specify axis argument
      ENH:  Add some simple image comparison metrics:  PSNR and NRMSE
      add new metrics to API
      TST: add tests for nrmse, psnr
      TST: fix broken test
      fix axis bug in multichannel case and add a corresponding test
      TST: add test for constent TV weighting across number of image dimensions for Chambolle algorithm
      update NRMSE docstrings and include run_module_suite in the corresponding test file
      TST: add test for invalid normalization type in NRMSE
      NRMSE, MSE, PSNR: make sure all inputs are float prior to the calculations.
      Fix the default dynamic_range in PSNR and warn if inputs don't meet the assumptions made
      fixed outdated import name _simple_metrics in __init__.py
      TST: add MSE, NRMSE test that would fail in case of int overflow
      psnr: update dynamic range calculation
      revert simple_metrics to np.float64 computations for all input types
      make suggested naming changes in simple_metrics.py
      use dtype=np.float64 in calls to np.mean

Guillaume Lemaitre (7):
      Added the Laplacian operator
      Generalize Laplce operator + testing
      Correct bug inside the test
      Add local geometric mean filter
      Update the documentation with spelling error
      Change the import of round function to be compatible with visual studio compiler
      Update the reference

Guillem Palou (4):
      Enforce SLIC superpixels connectivity
      Changed SLIC superpixels so that connectivity is enforced optionally.
      Added max_size as a parameter
      Changed tests to begin segment label at 1 instead of 0

Guillem Palou Visa (8):
      Changed coding style to be compliant with PEP8
      Fixing some code, minor revision
      Changed variable names for readbility, fixed PEP8 compliance
      fixing tests
      Merge branch 'slic-connectivity' of https://github.com/guillempalou/scikit-image into slic-connectivity
      Labels start at 0, for backward compatibility
      Fixed minor documentation issues
      Added deprecation warning

Gustav Larsson (2):
      Fixed crashing bug in slic.
      Added test for buffer overflow in slic.

Himanshu Mishra (23):
      Document the input image type
      Use astronaut image instead of lena in test_hog.py
      Raise a warning if the image is not grayscale
      Add expected_warnings to the tests using threshold_otsu
      Fix circular mask example in the user_guide
      Update plugins doc page
      Do not use lena in test_brief.py
      Remove use of lena in test_corner.py
      Remove test_otsu_lena_image
      Replace normalise kwarg in hog with transform_sqrt
      Modify tests for hog after renaming kwarg normalise
      Use astronaut instead of lena in test_match
      Raise error when single colored image is passed in threshold_otsu method
      DOC : Add comment for the use of strided output array for generic_filter in RAG
      Do not use lena in test_orb
      Add test for threshold_otsu on single color image
      Modify threshold_adaptive
      Correct example for threshold_adaptive
      Raise warning for 3D images in denoise_bilateral
      Fix test and doctest for multichannel kwarg in denoise_bilateral
      Raise error when normalise in not none in hog
      Add tests for warnings in denoise_bilateral
      Remove useless multichannel=True kwarg from tests

Holger Rapp (5):
      Fixed shutil.cp -> shutil.copy; the former doesn't exist on python2.6
      Added a more stable detection of opencv library by introducing a new file: _libimport.py
      Added cvFindChessboardCorners
      Added support for cvDrawChessboardCorners
      Merged with trunk

Horea Christian (12):
      User-specified output ratio
      Corrected indentation and spacing.
      Corrected indents (4space instead tab).
      Corrected indents Again.
      Added usage example.
      Refined syntax.
      Removed blank lines in between parameters.
      Renamed parameter
      Syntax and default variable.
      Adapted if statement to grid_shape=None
      Added test case for grid_shape parameter.
      Resolved faulty montage2d import.

Ivana Kajić (4):
      Added docu for gabor_filter. n_stds added to the parameter list.
      Added examples for Gabor filters.
      Updated Return section.
      More detailed docu. Modified examples.

Ivo Flipse (13):
      Update _hoghistogram.pyx
      Update _hog.py
      Made the indents line up with the (
      Extracted row_index and cell_row from the inner loop
      Moved more variables out of the loops and modified the docstring
      Moved more computations out of loops and added division from future
      I noticed that cy+cy and cx+cx was repeated in the inner-most loop as well
      Renamed x to rows and y to columns, though I might have still missed it in some spots.
      Converted orientation and magnitude to typed memory view. Also changed gradient_columns and gradient_rows, hopefully in the right way.
      I hadn't realized when I changed orientations to an array, that its actually just an int and the original code said range(orientations), so I changed it to np.arange(orientations). Hopefully the tests will now pass again
      I hadn't realized I was reusing cc in the loop, so it would crash after one iteration.
      Adding nogil to cell_hog
      Also added a nogil context manager to hoghistograms

JDWarner (14):
      Multispectral modifications applied to random walker.
      Added return_full_prob kwarg to solve call if pyamg not present.
      Added multispectral random walker test.
      Changes based on PR review recommendations: input format, scaling, and bugfix.
      sqrt(gradients) line removed
      PEP8 compliance, removed `scaling`, different data parsing.
      Documentation fix for `labels` input in random_walker.
      Docfix for img_as_float; appended myself to CONTRIBUTORS
      move `unpack_unique_verts` to Cython for speed
      fix: symmetric ambiguous dual-plane cases require special treatment
      ENH: Add `doctest_skip_parser` allowing conditional skipping of doctests
      FIX: New versions of FreeImage report software as "I", not "ImageMagick ..."
      FIX: Skip `viewer.show()` commands which hang Travis
      FIX: Specify field index for Python 2.6 compatible use of str.format()

JGoutin (2):
      Import fail if python is runned with -OO
      Update __init__.py

Jacopo (1):
      Updated docs for regionprops to reflect the dropping of regions with label value 0

Jaidev Deshpande (1):
      Fix typo in docstring

James Bergstra (19):
      ENH: initialize AffineTransform to identity when no args are provided
      FIX: added default identity matrix to SimilarityTransform to match AffineTransform
      Merge pull request #1 from ahojnnes/default-params-geometric
      FIX: corrected docstring for transform.warp
      FIX: transform.warp supports cval outside 0-1 range
      Added test case for warp() when cval out of clipping range
      ENH: moved code generating coords out of warp() to _warp_coords
      ENH: test_warp test case to make sure _warp_coords works for grey and rgb images
      ENH: renamed and docd _build_coords -> warp_coords
      ENH: test warp with data.lena instead of scipy.misc.lena
      ENH: remove call in tests to deprecated homography fn
      ENH: better docstring and test for warp_coords
      FIX: doctest of warp_coords
      ENH: warp_coords docstring
      ENH: stricter check on test_const_cval_out_of_range
      FIX: wordwrap to 78
      FIX: docstring mentions inverse instead of inverse_map
      FIX: configure just skimage logger on import
      ENH: default logger blocks messages from global handler

James Turner (7):
      JT: added PyFITS plugin, tests and data
      JT: various minor comments from Stefan
      JT: more minor comments from Stefan
      JT: fix an import when pyfits is missing
      Warn user when Cython can't be imported
      Allow for type of hdu.size changing in PyFITS 3.1
      Readability improvement suggested by tonysyu

Jan Margeta (2):
      Fixing MSVC build by adding appropriate typedefs for uint16_t
      Removing unnecessary typedefs for uint16_t in MSVC build

Jasper St. Pierre (1):
      transform: Remove some leftovers in __all__

Jeremy Metz (23):
      QH6228 Bug fix - remove offset from coords
      Added new convex_hull_image qhull test
      Removed trailing whitespaces
      Fixed PEP8 spaces
      Initial draft of Li thresholding (comments left in)
      Cleaned up comments, removed unnecessary vars
      Removed nbins input (from use with histogram input)
      PEP8 fixes, License file added, tests added
      Minor stylistic changes, removed lena test
      Moved contribution statement to CONTRIBUTORS.txt
      Merge branch 'master' into threshold_li
      Minor modification to docstring
      Updated exclude_border to not use min_distance
      Minor update to docstring
      Factor of 2 in remove added (due to footprint)
      Factor of 2 : PEP8 adjustment
      Added backwards compatibility for `exclude_border`
      Updated default value to be backwards compatible
      Updated test_peak for new exclude_border
      Reverted tests to boolean exclude_border
      Finally fixed issue:
      Added additional tests of exclude_border values
      Fix: missing colon on additional tests of exclude_border values

Joel Frederico (1):
      Update measure.regionprops.weighted_moments_central doc

Johannes L. Schönberger (10):
      Merge pull request #970 from stefanv/doc/peak_local_max
      Merge pull request #978 from jni/more-deprecations
      Merge pull request #930 from vighneshbirodkar/doh
      Merge pull request #973 from stefanv/coveralls_other
      Merge pull request #971 from stefanv/bug/views_are_copies
      Merge pull request #997 from stefanv/build/travis_numpy_py3
      Merge pull request #984 from JDWarner/randwalk_trivialcheck
      Merge pull request #1004 from jni/viewer-examples
      Merge pull request #992 from stefanv/warp_safe_output_shape
      Merge pull request #1007 from stefanv/label_refactor

Johannes Schönberger (79):
      added draw.polygon function with test cases
      replaced fixed output array size with dynamic lists
      improved doc string
      removed area function and removed old comment
      added example script for draw.polygon function
      removed geometry package from setup script
      added draw circle and ellipse functions
      improved doc string
      renamed function bresenham to line
      combined all draw functions in one example script
      changed interface of draw.polygon function for a more consistent usage
      improved color assignment of circle in shapes example script
      added explicit cython list type and changed creation of contiguous arrays
      DOC: update contributors
      added adaptive threshold
      fixed some typos
      added mean method to adaptive thresholding
      renamed adaptive_threshold to threshold_adaptive
      updated thresholding example script with more examples for adaptive thresholding
      clarified comment
      removed unused import
      adapted new function name for adaptive threshold in test cases
      updated test cases for adaptive thresholding
      added generic method to adaptive thresholding
      replaced unicode characters in doc string
      separated filter masks of mean filtering in adaptive thresholding
      improved doc string of adaptive threshold with more detailed description and example
      changed name of array variable
      restored old otsu threshold example script
      cython implementation of adaptive thresholding replaced with pure python version
      removed unnecessary cast step
      fixed typo in doc string
      fixed bug when sigma is set manually
      added regionprops function
      adapted to PEP8 whitespace conventions
      extended and improved description of properties in doc string
      moments and central moments no longer use separate functions
      removed old description of moments in doc string
      changed reference notes in doc string
      added test: labelled image must be of integer type
      most functionality of regionprops now pure python except for moments
      renamed regionprops source files for better separation between function name and source file
      fixed typo and improved example in doc string
      euler number now behaves like MATLAB implementation
      added test cases for regionprops
      more constistent way of determining ellipse parameters
      fix wrong indentation
      parameters of ellipse match Matlab results
      make all test cases of regionprops work
      update contribution for regionprops
      remove unused variables
      make matrix shape more readable in doc string
      change reference note in doc string
      reduce parameter choice of properties
      replace inline latex equations in doc string with plain text
      regionprops takes optional intensity images
      add example script for regionprops function to documentation
      add moments Cython extension to bento.info
      add perimeter measurement
      improve and restructure geomtric transformations
      remove old import
      add example to make_tform doc string
      update contributors
      fixe index errors and improve setup of some matrices
      apply PEP8 guideline
      remove inconsistent numeric correction and fix test case
      change API design and rename some functions
      change arguments of function estimate_transformation
      fix wrong exception type
      fix doc string formatting of function estimate_transformation
      redesign class interface
      fix inconsistent doc strings
      fix transformation union and add test case
      change interface of transformation merging
      add support for using transformation objects in warp function
      extend doc string example for geometric transformations
      add missing doc string for polynomial forward transformation
      geometric_transform can transform single coordinate tuple
      add short tutorial for geometric transformations

Johannes Schönberger (1426):
      Merge pull request #1 from stefanv/projection
      remove some blank lines
      fix bug in estimation of similarity transformation
      reimplement implicit parameter functionality of transformations
      combination of two transformations of the same type result in this type again
      fix scale initialization in implicit composition of similarity
      add doc for and restructure polynomial coefficients
      fix orientation of regionprops with correct quadrant determination
      add test case for correct quadrant determination
      fix and improve estimation of geometric transformation parameters
      fix incorrect comment
      fix and improve comments, doc strings, variable names for consistency reasons
      fix geometric transformation example after refactoring the module
      remove confusing comment
      handle composition of transformations with implicit parameters with __init__
      add mathematical description of estimation in doc strings
      fix typo
      change inverse_map parameter handling of warp function
      adapt geometric example script to new API and improve some expressions
      add test for shape of transformation matrix
      fix, improve and extend test cases related to geometric transformations
      add information about unit of angles to doc strings
      cast input of image label function
      fix num_peaks parameter bug in peak_local_max
      raise exception if matrix and implicit parameter arguments provided
      extend test cases for num_peaks parameter of peak_local_max
      complete default matrix initialization of geometric transforms
      add test case for polynomial transform initialization
      fix comment
      add local binary pattern texture analysis
      apply PEP8 guidelines
      merge contributors changes
      fix typos in doc string and comment
      add example script for local binary pattern
      move local binary pattern example script
      make local binary pattern example script use skimage sample images
      add plots to local binary pattern example
      rewrite local binary pattern in Cython for performance reasons
      fix bilinear interpolation when position outside of image
      fix test cases for local binary pattern
      improve doc string of local_binary_pattern function
      use new subplots function in example script of LBP
      Share bilinear interpolation function for other code
      Use shared bilinear interpolation function and improve performance.
      Wrap column
      Add doc string to bilinear_interpolation
      Rename matrix transform function and remove from pxd
      Add full-stop at end of short doc string description
      Remove unused imports and wrap text
      Fix typo in doc string
      Use numpy dtype objects instead of strings
      Make different methods of LBP more readable in Cython code
      Add doc string to Cython version of LBP
      Improve printed output of example
      Add plot labels
      Fix import bug due to rebase
      Add new package for shared code
      Move bilinear interpolation code to shared package
      Add integrate function to shared package
      Remove unused imports
      Add nearest neighbour interpolation
      Use predefined header files from Cython
      Refactor pnpoly function in Cython and add to shared package
      Fix indentation
      Add doc string to point in polygon functions
      Add empty line between functions
      Simplify code by removing duplicate loops
      Remove empty line
      Fix test cases of warps
      Merge pull request #253 from jaberg/warp_cval_outside_unit_interval
      add polygon approximation algorithm
      add example script for polygon approximation
      remove unused import and code from polygon approximation example
      remove colons in comments
      make loop condition more readable
      improved polygon approximation example script
      add subdivision of polygonal curves using B-Splines
      add note about size if resulting polygonal curves
      update polygon example with subdivision
      add test case for subdivision of polygons
      add some more comments to subdivision of polygons code
      make subdivision mask code more readable und debuggable
      fix wrong parameter name in doc string
      add test case for non-circular subdivision of polygons
      Add possibility to preserve ends and update tests
      Update subdivision example with preserve_ends parameter
      Add support for boolean dtype conversion
      Update test cases for boolean dtype conversion
      Add bool dtype range
      Fix typo in Cython compiler directives
      Optimize fast homography
      Merge pull request #262 from stefanv/doc_intersphinx
      Fix C-contiguous array bug in match template
      Fix doc string of texture detection
      Delete unnecessary file
      Fix shape format of arrays in doc strings
      Fix short description in doc string
      Fix parameter format in doc string of peak_local_max
      Capitalize parameter descriptions in draw package
      Fix examples of exposure package
      Capitalize parameter descriptions in geometric transforms
      Fix description of polygon example script
      Add deprecation warning in doc string to homography
      Refactor image warps
      Fix import error in geometric test cases
      Update radon transform with new warp function
      Fix doc string example of warp_coords
      Add function for image rotation
      Fix import of rotate function
      Fix rotate translation bug
      Fix translation bug in image rotate
      Rename parameter of image rotation
      Apply numpy doc style for deprecation warning
      Move image transform functions to _geometric file
      Add missing return type to matrix transform function
      Add function to clear border in binary images
      Refactor clear_border for better performance
      Move clear_border to segmentation package
      Use function pointer for different interpolation methods
      Simplify mode determination
      Add (bi-)cubic interpolation
      Add examples to doc string of clear_border
      Use explicit keyword for dtype
      Apply clipping also to fast cython implementation
      Fix example of estimate_transform
      Update doc string of _warp_fast for bicubic interpolation
      Add nearest mode for positions outside image
      Remove duplicate subtraction
      Add biquadratic interpolation
      Add image resize function
      Add new Coordinates property to regionprops
      Add example script for image labelling
      Improve visualization of labelling
      Update description with more detailed explanation of applied steps
      Explicitly define colormap
      Add parallel execution support
      Revert to non parallelized execution
      Fix decision whether to use warping or fast warping
      Refactor erosion and dilation for better performance
      Remove unused imports and add missing types
      Convert selem to uint8
      Improve doc string layout
      Add fast morphological operations for binary images
      Use numpy testing functions
      Fix support for predefined output array
      Fix for predefined output array
      Hide coefficient variable of geometric transforms
      Merge pull request #286 from TimSC/master
      Apply PEP8 and improve docs
      Speed up transformation of piecewise-affine
      Add missing doc string for inverse transform
      Implement inverse transformation for piecewise-affine
      Remove unnecessary params section in doc string
      Fix fast warping of images
      Add note about coordinates outside of mesh
      Fix inverse piecewise affine
      Add test cases for piecewise affine
      Fix typo in polygon example script doc
      Add example script for piecewise affine transform
      Add mesh points to plot
      Merge pull request #298 from tonysyu/edge-masking-tweak
      Add Coordinates property to property list
      Merge pull request #299 from ahojnnes/regionprops-coords
      Add function to build gaussian and laplacian pyramids
      Add missing references
      Rename file of pyramid functions
      Add tests for pyramid functions
      Fix wrong factor in pyramid_expand
      Fix singleton dimension for 1 pixel image
      Make pyramid functions generators
      Use consistent dtype for all levels of pyramid
      Rename factor parameters for better comprehensibility
      Add example script for image pyramids
      Add missing 0th layer of laplacian pyramid
      Update doc string return type of image pyramids
      Remove unnecessary C-contiguous flag
      Add test case for memory order
      Add dependencies to setup process
      Improve code layout
      Implement own dependency checking before installation
      Update pyramid example with longer description
      Use more readable variable name for image
      Add doc string to helper function _smooth
      Remove default parameter values from doc string
      Update short description of pyramid functions
      Add information about return dtype
      Use gaussian pyramid function for collection viewer example
      Update description of collection viewer example
      Fix bug in laplacian pyramid caused by renamed variables
      Update legacy comment
      Add another reference to laplacian pyramid
      Move comment to correct code
      Remove unused import
      Merge pull request #307 from kuantkid/inlinedisplay
      Make font-size of documentation smaller
      Merge pull request #306 from tonysyu/doc-negative-dtypes
      Remove scipy build dependency
      Merge pull request #312 from ahojnnes/dependency-fix
      Fix bug in grey erosion and dilation
      Let travis build bot run tests
      Fix bug in binary dilation and test case
      Add test case for non-square images
      Merge pull request #314 from tonysyu/doctest-fixes
      Merge pull request #316 from tonysyu/doc-img_as_bool
      Rename pyramid functions
      Fix typo
      Add more detailed description for pyramid functions
      Update name of pyramid function in pyramid example description
      Merge pull request #317 from tonysyu/testing-compatibility
      Add new doc theme
      Remove uncompressed bootstrap files
      Fix download link in navigation bar
      Rename scikits-image to scikit-image
      Rename scikits-learn to scikit-learn
      Merge pull request #338 from tonysyu/gallery-tweak
      Fix navigation docs link
      Extend CSS rule for all headings
      Fix navigation home page link
      Merge pull request #339 from tonysyu/doc-remove-feature-list
      Change domain name
      Remove unused fopenmp compile args
      Revert changes to release notes
      Add favicon for docs
      Add function to scale images
      Add test case for separate scale factors
      Rename scale function to rescale and scale variables
      Enhance short description of resize and rescale functions
      Fix doc version 0.7 link
      Fix bug in warp which caused 1 pixel images not to be clipped
      Full test coverage for image pyramids
      Full test coveragem for data package
      Full test coverage for canny filter
      Improve test coverage for ctmf
      Fix callable test in LPI filter
      Improve test coverage of LPI filter
      Fix polygon approximation for 0 or less tolerance
      Full test coverage for polygon approximation and subdivision
      Full test coverage for regionprops
      Full test coverage for structural similarity
      Improve test coverage of find_contours
      Remove obsolete license note as code has been refactored
      Remove redundant dtype conversion
      Merge pull request #355 from npinto/patch-1
      Merge pull request #354 from tonysyu/highlight-boundaries
      Merge pull request #353 from tonysyu/fix-test-setup
      Change Exception type for wrong input dims
      Merge pull request #357 from stefanv/doc_skimage_to_scikit-image
      Merge pull request #360 from tonysyu/fix-doctest-imports
      Add note about random test failure for some systems
      Split tests for different channel numbers into separate functions
      Add missing empty lines between test functions
      Add support for 3-dimensional image resizing
      Split 3d resize test function
      Fix failing perimeter test caused by bug in scipy
      Add hough transform peak detection
      Add note about range of angle values
      Fix typo
      Refactor non maximum suppression for faster computation
      Add missing support for num_peaks parameter
      Set constant value for maximum filter explicitly
      Add scharr edge filter
      Fix dtype and dim note for parameters in doc strings
      Fix reference links in doc strings
      Add reference for scharr operator
      Fix stylistic indentation
      Fix test case doc strings
      Add scharr filter to mask test cases
      Improve description of scharr and sobel edge filters
      Use common file for denoising filters
      Improve doc string format of TV denoise functions
      Add get_pixel function for 3D arrays
      Add bilateral denoising filter
      Restructure cdef declarations
      Use common file for denoising filters
      Refactor denoise tests and add tests for bilateral filter
      Add more detailed description of bilateral filter
      Add example script for bilateral denoising
      Wrap column
      Reorder subplots
      Fix description of denoising filters
      Fix PEP8 issues
      Fix typo in doc string of biliteral filter
      Rename tv_denoise to denoise_tv and deprecate
      Combine 2D and 3D bilateral filter in one function and rename sigma parameters
      Remove unused import
      Add note about range of values of sigma_range parameter
      Add Cython denoise source file to bento file
      Replace deprecated function call
      Add gabor filter function
      Fix meshgrid bug in gabor kernel and argument passing
      Add test cases for gabor filter
      Add example script for gabor filter
      Remove unnecessary kernel description for filtering
      Fix ordering of return values
      Fix hough transform return value ordering in doc string
      Reduce to one threshold paramater
      Use center of mass for peaks with equal accumulator values
      Fix threshold default value description
      Merge branch 'rank_filters' of https://github.com/odebeir/scikits-image into rank-filters
      Improve overall code style
      Improve code layout of kernel functions
      Change type of pop variable from float to Py_ssize_t
      Clean Readme
      Wrap lines
      Remove duplicate contiguous conversion
      Fix, improve and extend histogram test cases
      Do not ignore exceptions in core functions
      Fix, improve and extend test cases
      Use single file for test cases and rename file
      Change data type of pop variable to float
      Fix typos and spelling of readme
      Fix spelling of X-bit
      PEP8 fixes in doc string examples
      Use numpy reference style
      Improve description of long rank filter example
      Use common spelling for greylevel
      PEP8 fixes for long rank filter example
      Merge pull request #348 from odebeir/rank_filters
      Apply numpy reference style to rank filter descriptions
      Merge pull request #374 from tonysyu/add-rank-to-bentoinfo
      Add example for hough_peaks function
      Merge pull request #380 from tonysyu/mark-boundaries-regression
      Add title to hough peak detection example plot
      Merge pull request #379 from cpsaltis/fix_rotate
      Fix dependency check in setup
      Move harris file to a combined interest point file
      Refactor harris corner detection, so it returns harris response image
      Add Moravec interest point detection
      Rename harris test case file
      Make moravec standalone cython function and fix doc string
      Fix return types in doc string of interest point detectors
      Add reference to moravec operator
      Rename window size parameter of moravec operator
      Fix missing sigma for gaussian filter in harris operator
      Fix harris corner bug and add alternative corner measure param
      Add comment to explain meaning of structure matrix of harris
      Add Shi-Tomasi corner detector and fix bug in auto-correlation matrix
      Update test cases for shi-tomasi and moravec operator
      Fix contiguous bug in moravec operator
      Add example to doc string of shi-tomasi
      Add some more references to interest point functions
      Add missing doc string return section
      Improve code layout
      Rename corner operators and file
      Add Kitchen and Rosenfeld corner detector
      Use sobel for image derivatives
      Add more detailed description of corner detectors in doc strings
      Add foerstner corner detector
      Fix short description of moravec corner detector
      Let foerstner return both corner measures
      Add subpixel corner detection
      Add references for subpixel corner detection
      Add test case for subpixel corner detection
      Update examples of corner detection in doc string
      Refactor corner detection example
      Fix doc string for multiple return values
      Fix typo in doc string
      Add comments and rename variables for better readability
      Remove unused array
      Fix confused parameters in doc string of peak_local_max
      Add function to detect corner peaks as a wrapper to peak_local_max
      Extend example of corner_peaks to highlight difference to peak_local_max
      Add test case for corner_peaks function
      UmfPack warning is now only raised if random walker is explicitly called
      Fix some doc string formatting issues of random walker
      Merge pull request #392 from blink1073/fast-import
      Hide functools and imp packages in global skimage namespace
      Fix double space in warning message
      Fix recursion problem with nose test suite
      Merge pull request #394 from tonysyu/circular-import-fix
      Add default value for polynomial order
      Add test case for default value of polynomial order
      Fix typo bug in perimeter function
      Fix convolution bug in perimeter function
      Update perimeter test case which was correct for buggy previous version
      Add note about default polynomial order to estimate_transform
      Merge pull request #404 from keflavich/clearer_warp_example
      Fix PEP8 issues and doc string formatting
      Improve total-varation denoising algorithm
      Fix test cases for TV denoising
      Remove old implementation if TV filter
      Update denoising example for new TV implementation
      Replace manual array with builting Cython buffer indexing
      Rename denoise_tv to denoise_tv_bregman
      Rename denoising cython source file
      Fix typo in denoise_tv_bregman doc string
      Rename test functions for bregman algorithm
      Add original (Chambolle) implementation of TV denoising
      Update denoising example
      Fix bugs caused by renaming functions
      Add test cases for chambolle tv denoising implementation
      Add test case for multi-channel chambolle denoising
      Add note about multichannel parameter
      Include more information about reference papers
      Merge pull request #384 from andersbll/master
      Merge pull request #410 from sciunto/circle
      Fix typo in doc string
      Use char type for circle_perimeter method for speedup
      Use same cython directives for whole source file
      Fix PEP8 issues
      Use char type for circle_perimeter method for speedup
      Use same cython directives for whole source file
      Fix PEP8 issues
      Fix PEP8 issues
      Add empty lines between functions
      Wrap line and group imports
      Make some stylistic code layout changes
      Merge pull request #415 from tonysyu/update-bento-build
      Merge pull request #419 from JDWarner/fix_noisy_square_seed
      Set random seed value to ensure tests always pass
      Set random seed value for denoising tests
      Merge pull request #422 from colincsl/hog_visualisation
      Merge pull request #423 from cgohlke/patch-1
      Fix wrong include paths for shared subpackage
      Merge pull request #432 from thouis/sse_ctmf
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Rename Py_ssize_t to ssize_t for consistency within skimage
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Change type to ssize_t for all index and size variables
      Use np.intp rather than ssize_t for ndarrays
      Replace ssize_t with Py_ssize_t globally
      Fix index type errors in draw module
      Remove print statement
      Fix index type error in SLIC function
      Fix index type error in GLCM
      Improve performance and type usage for GLCM loop
      Fix integer type bugs in feature detection algorithms
      Improve code layout of probabilistic hough
      Remove unused CAPI numpy import and extend cython directives
      Remove unused imports
      Make list of zip in Python 3 explicitely
      Wrap lines
      Add note about proper use of Py_ssize_t
      Globally change np to cnp if cimported
      Fix accidentially changed doc strings
      Merge pull request #436 from sciunto/docstring
      Merge pull request #417 from stefanv/msvc_build_fix
      Remove duplicate return statement
      Remove deprecated homography function
      Remove deprecated import of hompgraphy
      Merge pull request #439 from tonysyu/fix-equalize
      Remove deprecated tests for homography
      Merge pull request #443 from stefanv/hough_brighter_color
      Merge pull request #445 from tonysyu/doc-tweaks
      Merge pull request #447 from stefanv/ndimage_dtype_32
      Change data type of index variables
      Improve performance and usage of data types of hough circle transformation
      Remove unnecessary statement breaks
      Change color of circles for better visibility
      Merge pull request #446 from jni/morphology-functions
      Improve performance of normalization
      Merge pull request #452 from stefanv/jni-morphology-functions
      Add release notes for version 0.8
      Update version number for 0.8
      Update docversion for 0.8
      Fix release instruction for making docs
      Increase version number to 0.9
      Merge pull request #453 from cgohlke/patch-1
      Add mailing list adresses to release instructions
      Add release instruction to update dev docs
      Fix vectorized_ops for bento
      Remove colon as multi-line comment symbol
      Merge pull request #456 from cgohlke/patch-2
      Change deprecated decorator to display warning in doc string
      Remove duplicate code for alternative function
      Use double instead of single quotes for doc string
      Make deprecation warning in doc string bold
      Add API documentation to latex docs
      Increase list-depth for nested quotes etc.
      Change latex doc settings
      Fix some doc strings for latex docs
      Re-enable plots in doc strings and generate PDF images for them
      Explicitly set latex font size and paper size
      Use plot_directive from matplotlib.sphinxext if available
      Copy some useful latex settings from NumPy
      Fix make.bat
      Avoid duplicate python module index
      Do not set paper size to allow manual setting through make
      Fix parameter list format error of rotate
      Replace scipy.ndimage routine with skimage.ndimage.rotate
      Use nearest neighbour interpolation to avoid different labels
      Remove unused import
      Replace rotate function of local binary pattern example with builtin skimage function
      Explicitly import local_binary_pattern function
      Replace ndimage.zoom function in radon transform example with builtin skimage function
      Merge pull request #467 from malcolmreynolds/fix-num-peaks-typo
      Fix reference syntax for corner detectors
      Merge pull request #475 from sciunto/docstring
      Merge pull request #480 from sciunto/hl_review
      Make sure non-zero values are treated as True
      Add test case for different dtypes of structuring element
      Merge pull request #483 from almarklein/windowserror
      Merge pull request #482 from thouis/master
      Merge pull request #486 from yangzetian/peak-local-max-indices-nd
      Update release instructions for new robots.txt file
      Rename sigma parameters by adding an underscore as separator
      Merge pull request #2 from tonysyu/gabor
      Merge pull request #3 from tonysyu/gabor
      Merge pull request #493 from ChrisBeaumont/vector_integral
      Inline helper functions of label
      Add option to return number of labels
      Add test case for number of labels
      Rename development guidelines, so they are shown when opening a new PR
      Make return type of skeletonize function boolean
      Reformat code to comply with PEP8
      Remove unnecessary conversion step to numpy array
      Remove duplicate import statement
      Merge pull request #525 from stefanv/felzenwalb_32
      Merge pull request #535 from JDWarner/fix_mcp
      Pre-build edge filter weights and use views
      Remove outdated author in doc string
      Fix output array dimensions in doc string
      Fix hough line examples
      Add short description to edge filter example
      Rename mean filter example and improve description
      Use numpy dtype rather than string
      Fix inconsistencies in examples and many more improvements
      Let Travis run all examples
      Escape dollar-sign in makefile
      Add matplotlib to Travis dependencies
      Set matplotlib backend to Agg
      Use current matplotlib version Python 3
      Escape greather-than symbol
      Escape double-colon symbol
      Change python version test
      Fix wrong escaping
      Add missing python version to pip
      Remove escaping
      Change invalid install order
      Move example runner loop to Travis config
      Unescape dollar sign
      Add better description for example runner
      Fix print statement for Python 3
      Add missing future import
      Add missing trailing space to print statement
      Install matplotlib after freeimage
      Use png rather than jpg for chelsea image
      Use png rather than jpg for ihc image
      Fix image filename extension from .jpg to .png
      Convert to ubyte dtype explicitly for python 3
      Fix dtype bug in color tests for python 3
      Convert to ubyte dtype explicitly
      Deprecate bresenham function
      Replace all occurences of bresenham with line
      Pass error return code of example scripts and exit
      Add __all__ to color package
      Add __all__ to exposure package
      Explicitly import color functions
      Add __all__ to data package
      Add __all__ to draw package
      Add __all__ to feature package
      Add __all__ to filter package
      Add __all__ to rank filter package
      Add __all__ to graph package
      Add __all__ to measure package
      Add __all__ to morphology package
      Add __all__ to segmentation package
      Add __all__ to transform package
      Add __all__ to util package
      Fix selem test cases
      Fix hough transform test cases
      Add line estimation model
      Add RANSAC algorithm
      Add circle estimator model and fix some other bugs
      Fix doc string of ransac function
      Add ellipse estimator model
      Rename variables for better readability
      Add todo comment
      Add example to ransac function
      Add imports of fit to subpackage
      Fix example of ransac
      Reduce default number of max trials
      Add analytical jacobian for ellipse model for 100x speedup
      Reduce number of function calls for speedup
      Calculate initial guess for closest point on ellipse for speedup
      Replace numpy with math for scalar functions and remove Dfun from ellipse residuals for speedup
      Fix typos
      Add input data shape test to avoid out of memory Jacobians
      Fix bug in predict_x of line model
      Return combined xy-coordinate array with arbitrary dimensions in predict_xy
      Fix bug in ellipse model estimation due to changed predict_xy output
      Add test cases for line model
      Add test cases for circle model
      Add test case for ransac algorithm
      Replace triple-single-quotes with triple-double-quotes for doc strings
      Add is_degenerate to geometric transforms
      Add residuals method to geometric transforms
      Change ransac function to accept multiple input data arrays
      Fix is_degenerate function to accept src and dst arrays separately
      Add new stop criteria to ransac and improve doc string
      Add example for ransac and geometric transforms
      Add test case for ransac applied to geometric transform
      Remove trailing brackets
      Add is_degenerate to description of ransac iteration
      Replace is_degenerate with is_model_valid and is_data_valid in ransac
      Remove is_degenerate methods from estimation models
      Replace is_degenerate tests with is_data_valid and is_model_valid tests
      Add reference for RANSAC
      Add example matching script
      Remove unused shear argument
      Remove extra space
      Clarify SSD description
      Add comment to explain commented out code
      Use double quotes for code in doc string
      Add missing double colons for equations
      Add missing double quotes for equations and code
      Update doc string of ransac with new validation functions
      Add title of wikipedia reference
      Add test case for underdetermined LineModel estimation
      Add test cases for residuals
      Readd Cellprofiler license notes to edges.py
      Fix last remaining PEP8 errors
      Use boolean mask for inlier return value of RANSAC
      Wrap lines
      Add simple RANSAC example script
      Set random seed to avoid random test failures
      Fix spelling
      Add python code checkers to travis build process
      Replace pep8 and pyflakes with flake8
      Use mirrors for flake8 installation
      Add exit-zero directive to flake8 command
      Merge pull request #563 from JDWarner/docfix_label
      Add short example for transforms to warp function
      Fix parameter description of warp function
      Fix parameter descriptions of warp functions
      Add examples to doc strings of resize, rotate and rescale functions
      Fix spelling error
      Fix incomplete options for mode of _warp_fast function
      Wrap lines
      Fix doc string injection of deprecated wrapper
      Fix invalid index of reference
      Escape latex math symbols
      Add filter.rank to __all__ of filter package
      Fix indentation of ransac doc string example
      Fix styling of math and blockquote
      Fix doc string of threshold_adaptive
      Remove bold font from dt tags
      Fix quick search description
      Update installation instructions
      Remove unnecessary bscript
      Implement faster version of circle and ellipse drawing
      Capitalize parameter description
      Add example to line drawing function
      Add example to polygon drawing function
      Add example to circle_perimeter drawing function
      Add example to ellipse_perimeter drawing function
      Add example to ellipse drawing function
      Add example to circle drawing function
      Add example to set_color drawing function
      Merge pull request #532 from ankit-maverick/deprecate_is_rgb
      Merge pull request #510 from sciunto/ellipsev2
      Set matplotlib backend globally
      Create matplotlib config directory
      Set matplotlib qt backend
      Avoid import of matplotlib QT lib if QT is not installed
      Use None instead of 'none' for qt backend
      Fix setting of QT_API environment variable
      Fix python 3 print statement in example script
      Fix window size of sub-pixel corner detection
      Fix matching example script
      Merge pull request #572 from sciunto/docandco
      Merge pull request #578 from josteinbf/radon-test-refactor
      Merge pull request #576 from alexis-mignon/master
      Merge pull request #570 from stefanv/bento_hookfile
      Merge pull request #583 from josteinbf/radon-example-axes-fix
      Merge pull request #505 from ankit-maverick/gammaCorrect
      Merge pull request #585 from ankit-maverick/typos
      Merge pull request #586 from sciunto/print
      Merge pull request #587 from sciunto/import
      Merge pull request #589 from sciunto/configparser
      Refactor contribution workflow
      Change branch name for consistency across guidelines
      Fix typo
      Add additional advice about rebasing
      Use long parameter version --force rather than -f
      Merge pull request #4 from tonysyu/rebase-workflow-edits
      Merge pull request #598 from sciunto/fix_import
      Merge pull request #597 from sciunto/hough_ellipse
      Fix division by zero bug in corner detectors
      Merge pull request #604 from JDWarner/fix_sitk_tests
      Merge pull request #621 from tonysyu/fix/rank-dtype-compatibility
      Merge pull request #617 from JDWarner/memview_transform
      Merge pull request #625 from JDWarner/random_noise
      Outsource inner brief loop into cython file
      Outsource inner brief loop into cython file
      Merge pull request #511 from ankit-maverick/resample
      Refactor local block functions and add additional functionality.
      Fix import namespace
      Fix doc string examples
      Update test cases for block functions
      Update doc string for new functionality
      Rename block_* functions to local_*
      Fix deprecated function name referneces in doc strings
      Use pad function and add option to define cval
      Merge pull request #643 from stefanv/bento_fixes
      Let travis-ci check bento.info
      Merge pull request #648 from stefanv/ransac_faulty_cast
      Merge pull request #652 from josteinbf/local-binary-pattern-example
      Refactor rank filter package for consistent naming
      Use typed memoryviews in rank filter package
      Improve code layout and styling
      Use typed memoryviews for diff array
      Fix styling of some comments
      Fix wrong module name
      Capitalize parameter description in doc string
      Extend bitdepth from 12bit to full 16bit
      Fix attack-release border initialization
      Fix test cases for full 16bit support
      Fix README for full 16bit support
      Use missing memoryviews for bilateral kernels
      Improve description of bilateral filter
      Add test case for bilateral filters
      Improve indentation
      Cast to int for python 3
      Update bento.info for refactored rank package
      Log warning when bitdepth > 10
      Refactor rank filter package as combined implementation for 8- and 16-bit
      Update bento config
      Improve entropy example
      Improve mean example
      Update entropy example with improved matplotlib usage
      Improve long rank filter example
      Update bitdepth warning
      Replace log warning with UserWarning
      Rename mean_subtraction, morph_contr_enh to subtract_mean and enhance_contrast
      Use consistent description of output image
      Append percentile, bilateral function name part
      Add TODO.txt
      Add note about TODO.txt to release instructions
      Deprecate old rank filter functions
      Replace deprecated function calls in examples
      Add option to return double output image for rank filters
      Replace all float dtypes with double
      Add test for output dtype of entropy
      Merge pull request #546 from jni/3d-slic
      Rename and combine local_* functions to block_reduce
      Add short doc string example
      Merge pull request #591 from ankit-maverick/brief
      Merge pull request #660 from RONNCC/add_to_depends
      Merge pull request #561 from RONNCC/fix_filter_doc_gaussian
      Merge pull request #659 from RONNCC/add_bento_to_docs
      Replace deprecated import of _local_func with block_reduce
      Merge pull request #653 from iNand/master
      Fix typo in doc string
      Wrap line
      Add missing space in parameter defintion in doc string
      Add missing p1 parameter to percentile and threshold functions
      Add specific branch for the percentile p0 = 0 case
      Fix accidential variable naming confusion
      Integrate new test cases from @odebeir and fix old function names of dilate and erode
      Fix invalid previous fix for percentile edge case
      Fix loop ranges of rank filter kernel functions
      Fix Otsu rank filter kernel loop range
      Let travis also run long example scripts
      Add short comment for travis flake8
      Fix image data type issues in rank filter example
      Merge pull request #641 from sciunto/example_ht
      Add cached_property decorator
      Refactor regionprops
      Remove cached_property class from shared utils
      Add note about backwards compatibility to TODO.txt
      Remove legacy doc strings
      Use asbolute import for Cython lib
      Remove unused cached_property from _shared.utils namespace
      Improve fromatting of regionprops doc string
      Hide cached_property and caching flag
      Merge pull request #669 from ankit-maverick/octagon
      Fix radon example script
      Merge pull request #684 from sciunto/doc_fix_url
      Fix missing label property description and backwards compatibility
      Fix bugs in censure keypoint detector and improve code
      Fix several bugs in DoB method and improve overall code quality
      Fix censure example and fix some minor issues
      Merge pull request #668 from ankit-maverick/censure
      Improve moments code and use typed memoryviews
      Make moments functions public, rename, add tests
      Sort functions in alphabetical order
      Add test cases for moments functions
      Fix PEP8 issues and typos
      Use typed memoryviews in draw package
      Use typed memoryviews in feature package
      Use typed memoryviews in filter package
      Fix reference of corner_foerstner
      Fix missing conversion of typed memoryview to numpy array
      Fix missing conversion of typed memoryview to numpy array
      Improve tests of graph package
      Use typed memoryviews in measure package
      Use typed memoryviews in morphology package
      Use typed memoryviews in transform package
      Fix bugs in usage of typed memoryviews
      Convert to int for Py_ssize_t
      Merge pull request #700 from stefanv/remove_logger
      Merge pull request #702 from stefanv/remove_image_wrap
      Merge pull request #720 from michaelaye/patch-1
      Remove unnecessary mirror option
      Improve SLIC
      Reduce memory footprint
      Remove unnecessary parameters and update doc string
      Remove legacy comment
      Fix dtype bug
      Fix bug in window extent determination
      Reorder variable declarations
      Improve documentation of sigma
      Revert multichannel magic and improve parameter docs
      Use absolute imports
      Add missing rgb2lab import
      Merge pull request #722 from stefanv/shape_step
      Make image C-contiguous
      Rename clusters to segments
      Add optional description to parameters
      Fix indentation
      Change default value of sigma
      Merge pull request #686 from radioxoma/yen-threshold
      Merge pull request #704 from alexis-mignon/master
      Merge pull request #630 from JDWarner/improve_noise
      Merge pull request #727 from jni/fix-small
      Revert default value of maximum iterations
      Merge pull request #5 from jni/slic-speed-fix-dims
      Raise warning for sigma parameter and add to TODO for next release
      Merge pull request #6 from jni/slic-speed-spacing
      Merge pull request #731 from ankit-maverick/SimilarityTransform
      Document default value API change
      Merge pull request #7 from jni/slic-cython-docstring
      Merge pull request #733 from riaanvddool/contribution-riaanvddool
      Define 1d arrays as contiguous
      Scale sigma only in scalar case
      Fix doc string of warp functions
      Fix various doc strings using the warp function
      Merge pull request #746 from luispedro/master
      Some improvements for perimeter
      Add support for unsigned integer label images
      Raise error for non 2-D images
      Fix dtype bug in perimeter
      Hide experimental GSoC functions for 0.9 release
      Remove call to deprecated is_gray function
      Misc doc string fixes
      Only clip for ndimage.map_coordinates case
      Add note about internal usage of transformation matrix
      Add support for matrix as inverse_map
      Improve example of warp function to show all available options
      Fix deprecated function name in example
      Temporarily remove censure keypoints example
      Merge pull request #752 from sciunto/pep
      Merge pull request #755 from jni/rename-relabel
      Merge pull request #709 from sciunto/wu
      Merge pull request #762 from jni/update-slic-example
      Merge pull request #764 from jni/example-py3
      Merge pull request #767 from cgohlke/patch-1
      Merge pull request #771 from cgohlke/patch-4
      Fix euler number bug for scipy-0.13
      Add missing perimeter function to __all__
      Merge pull request #768 from cgohlke/patch-2
      Fix numpy 1.8 bug
      Split grey and binary erosion tests in two files
      Fix overflow error
      Reduce number of specializations
      Remove unused import
      Add support older numpy versions
      Deprecate out parameter
      Improve shape plot
      Use libc.math.atan2
      Miscellaneous fixes and improvements
      Merge pull request #776 from tonysyu/update-api-doc
      Merge pull request #772 from JDWarner/fix_poisson_noise
      Merge pull request #775 from JDWarner/spacing_random_walker
      Add logger deprecation to todo list
      Merge pull request #780 from JDWarner/spacing_marching_cubes
      Merge pull request #781 from stefanv/gh_stable_doc_versions
      Merge pull request #782 from sciunto/pep8
      Merge pull request #778 from stefanv/binary_morph_out
      Remove duplicate set_color function
      Add test case for set_color function
      Re-add missing parts
      Merge pull request #624 from mkcor/filter_median_dtypes
      Remove deprecated is_local_maximum function
      Add deprecated is_rgb and is_gray functions to todo list
      Remove deprecated tv_denoise function
      Remove deprecated hough_* functions
      Re-add tests
      Remove print statement
      Rename weighted central moments property
      Fix dtype error on 32bit systems as convolve is only implemented for 32bit integers
      Change data type of convolution to more general uint
      Speed up memory views in local_binary_pattern
      Speed up memory views in line drawing function
      Speed up memory views in skeletonize function
      Speed up memory views in watershed function
      Fix bug in homography detection
      Explicitly cast image shape to integer value
      Add some more comments
      Re-enable clipping which was accidentally removed
      Do not derive piecewise affine from projective transform
      Merge pull request #600 from tonysyu/novice-package
      Merge pull request #807 from josteinbf/regionprops-label-sequence
      Add rank filter tests as data dir
      Fix 16bit rank filter test in python 3.x
      Merge pull request #801 from stefanv/six_external
      Add make instruction for doctest
      Let travis run doctests
      Fix corner doc tests
      Fix corner_subpix function for corners near border
      Add example for corner_subpix function
      Fix doctest of remove_small_objects
      Fix watershed doctest
      Fix hough_line_peaks doctest
      Fix block_reduce doctest
      Fix RANSAC doctest
      Use NORMALIZE_WHITESPACE directive for block_reduce doctest
      Fix montage2d doctest
      Remove plotter from doctest and use comparison to np.histogram instead
      Fix precision of output in hough_line_peaks doctest
      Install nose from PyPI
      Remove unnecessary coverage directive
      Install PyQt4 for doctests
      Install Cython from apt
      Fix PyQt4 for python 3
      Install xserver
      Install Cython via pip
      Use xvfb rather than xorg
      Do not change into a separate directory for testing
      Build extensions in-place
      Fix duplicate percentile names rank namespace
      Fix duplicate find_contours names in namespace
      Remove numpy prefix from pad function examples
      Skip ransac doctests
      Only build in-place
      Update bento.info
      Fix novice doctests
      Fix multi image doctest
      Explicitly cast window extent to integer for python 3
      Fix novice doctest
      Fix corner_foerstner doctest
      Fix python 3 syntax error
      Also run doc tests for files with a leading underscore
      Temporarily disable brief test cases and ignore test files with a leading underscore
      Fix rank order doc test
      Fix marching_cubes doctest
      Fix regionprops doctest
      Fix medial_axis and skeletonize doctests
      Fix geometric doctests
      Fix downscale_local_means doctest
      Fix hough_line_peaks doctest
      Fix data type issue in python 3
      Improve usage of doctest directives
      Use a better way to comment out the brief examples temporarily
      Revert ellipsis to skip directive
      Use travis-ci python config
      Discard pip installation and install matplotlib from PyPI
      Simplify travis config
      Touch non-existent matplotlibrc
      Fix matplotlibrc configuration
      Add python 2.6 and 3.3 environments
      Remove python 2.6 and 3.3
      Cast label_field to integer dtype for float intput
      Fix label2rgb bug
      Fix misspelled sections
      Fix doc string syntax errors
      Reduce runtime of daisy visualization test case
      Reduce runtime of radon test cases
      Explicitly set sigma value to 0 to avoid warning
      Skip show commands for doc tests
      Add note about disabled doctests
      Enable doc tests of ransac function
      Add doctest function to skimage
      Use skimage test functions instead of nose commandline
      Add --exe directive to doc test
      Return correct exit code for tests
      Make sure stdlib io is not shadowed by skimage.io
      Skip failing ransac doctest
      Reduce runtime of long doctests
      Fix indentation
      Use explicit interger division
      Merge pull request #819 from stefanv/coverall_support
      Merge pull request #804 from tonysyu/viewer-color-hist-fix
      Remove duplicate greycomatrix function definitions
      Add test case for *_boundaries functions
      Test invalid inplace operation
      Add missing deprecation removal to todo
      Add missing deprecation removal to todo
      Add test case for union of heterogenous tform types
      Add test cases for abstract geometric base class
      Test invalid input
      Test scale determination for 0 rotation
      Add missing test coverage of polynomial tform
      Add test case for estimate_transform
      Add matrix_transform function to namespace
      Deprecate reverse_map parameter and add to future removal list
      Add test case for invalid image dimensions
      Fix bug in inverse warping and add test case
      Rename test to enable hidden line model test
      Increase test coverage of regionprops
      Add test for invalid parameter combination
      Add test case for eps corner_harris method
      Add test for mask output of corner_peaks
      Add missing removal of deprecated function to todo
      Add test for negative input image
      Add test for neggative gamma parameter
      Add test case for explicit use of radii and sigmas parameters
      Fix error in line model test
      Merge pull request #827 from stefanv/coveralls_nomain
      Merge pull request #821 from JDWarner/add_conditional_doctest_decorator
      Fix some performance regressions in Cython implementation of FAST
      Fix some performance regressions in Cython implementation of FAST
      Fix invalid circle coordinates
      Fix some performance regressions in Cython implementation of FAST
      Fix some performance regressions in Cython implementation of FAST
      Fix invalid circle coordinates
      Transform corner_fast_orientations for application to arbitrary corner types
      Move FAST corner function to other corner functions
      Make structure tensor function public and add function to compute hessian matrix
      Add missing imports
      Add test cases for structure tensor and hessian matrix functions
      Add functions to compute eigen values of structure tensor and hessian matrix
      Add examples for structure tensor and hessian matrix functions
      Add test cases for eigen value functions of structure tensor and hessian matrix
      Miscellaneous fixes and improvements
      Finish create_keypoint_recarray helper function
      Improve BRIEF and various tweaks
      Improve FAST example output
      Add comments to FAST implementation
      Improve corner_orientations example output
      Implement object oriented interface for BRIEF
      Reactive BRIEF test cases
      Add test case for border keypoints
      Improve example output of corner_moravec
      Implement CenSurE detector in object oriented interface
      Refactor ORB
      Refactor match_descriptors and fix small bugs in ORB
      Use dynamic pad-width for arbitrary mask shapes
      Fix test cases
      Update bento.info
      Fix legacy import
      Update BRIEF example
      Add example to ORB doc string
      Add note about usage of ORB
      Add example to doc string of CenSurE
      Fix corner doctests
      Add test cases for feature utilities
      Fix imports for Python 3
      Add test for unsupported mode
      Add short description to ORB example
      Add convenience function for plotting matches
      Use new plot_matches function for plot_matching example script
      Remove prints in test function
      Fix wrong parameter name in doc string
      Convert shape to list to allow for in place modification
      Add example script for BRIEF
      Add example script for CenSurE
      Improve layout of gabor example plot
      Fix and improve circular hough hough transform example script
      Fix overlapping text
      Fix overlapping text
      Improve local otsu example script
      Fix overlapping xtick labels
      Show hidden text
      Improve ssim example
      Attempt to fix Python 3 error
      Add tests for plot_matches
      Fix parameter list in doc string of plot_matches
      Combine match indices in one array
      Recover hidden test cases caused by duplicate function names
      Speedup corner orientations
      Fix plot_matching example
      Add missing plot directive
      Fix description
      Fix typo
      Add missing optional description for kwargs
      Fix bug in array padding, template matching
      Fix doctests
      Update arraypad tests
      Add support for 3-D template matching
      Reuse 2D sum in 3D sum
      Add tests for 3-D template matching
      Add tests for wrong input shapes
      Move inlined function definitions to header for actual inlining
      Fix doc tests
      Avoid temporary array allocations
      Add missing bracket in doc tests
      Re-order input checks
      Fix test case
      Fix ndim test
      Improve description of match_template
      Refactor final slicing
      Add test for reflected padding
      Fix and improve doc string
      Merge pull request #852 from jni/rgrid-int-steps
      Merge pull request #844 from odebeir/add_rank_sum
      Improve doc strings of generic rank filters
      Improve doc strings of bilateral rank filters
      Improve doc strings of percentile rank filters
      Fix pop_bilateral doc test
      Clarify that it is the local histogram
      Readd note about overflow
      Use 2 empty lines between functions
      Merge pull request #860 from jaidevd/jd_skimeasure_typo
      Add note about dtype overflow
      Fix input test of template matching
      Merge pull request #843 from tonysyu/refactor/io
      Merge pull request #831 from astrofrog/use-cythonize
      Merge pull request #857 from guillempalou/slic-connectivity
      Save feature information in attributes for consistency
      Use public attribute for parameter values
      Use public attribute for parameter values
      Add tests for deprecated attributes
      Remove wrong comment location
      Fix typo
      Use None as default for estimated params
      Use consistent params_ attribute across all estimators
      Fix match tests
      Fix geometric example
      Fix BRIEF doc test
      Get rid of trailing underscore for params attribute
      Document params attribute for each class
      Fix and improve BRIEF example
      Improve ORB example
      Improve BRIEF doc string
      Get rid of trailing underscores
      Fix parameter notation in doc string
      Rename CenSurE to CENSURE
      Fix match tests
      Fix ORB example script
      Merge pull request #870 from stefanv/bugs/windows-round
      Merge pull request #887 from stefanv/docs/gh-pages
      Implement missing max_distance parameter
      Merge pull request #889 from sciunto/tuto
      Remove six copy in _shared as it is a required dependency
      Fix peak_local_max doc string
      Add test case for 3D images
      Add test case for 4D images
      Merge pull request #911 from cgohlke/patch-5
      Merge pull request #910 from cgohlke/patch-4
      Merge pull request #908 from cgohlke/patch-2
      Merge pull request #909 from cgohlke/patch-3
      Merge pull request #912 from cgohlke/patch-6
      Fix warp doctest and example
      Fix regular grid doctest
      Always Skip doctests which depend on Qt and are in the class doc string
      Improve function name in warp example
      Move phase unwrapping algorithms to restoration submodule
      Move denoise functions to restoration submodule
      Fix restoration example imports
      Update bento
      Add denoise deprecation removal to TODO.txt
      Merge pull request #882 from JDWarner/correct_mesh_orientation
      Extract changes by @synesthesiam in PR #797
      Upgrade numpydoc to version 0.4
      Merge pull request #947 from stefanv/build/contribs
      Merge pull request #948 from stefanv/bug/plot_matches-cmap-interp
      Merge pull request #957 from sciunto/fixdoc
      Merge pull request #946 from synesthesiam/master
      Warn about matplotlib backend in skimage.viewer
      Merge pull request #960 from sciunto/doc_enhan
      Temporarily remove video plugin
      Update subpackage descriptions
      Homogenize subpackage description
      Document removed skimage.io.video in API changes
      Merge pull request #956 from vighneshbirodkar/issue945
      Remove deprecated skimage.exposure.exposure
      Remove deprecated skimage.morphology.greyscale_*
      Remove deprecated skimage.segmentation.visualize_boundaries
      Remove legacy todo item for BRIEF, which has been addressed
      Remove deprecated skimage.filter.median_filter
      Remove deprecated logger function in skimage/__init__.py
      Remove backwards-compatability of skimage.measure.regionprops
      Use new regionprops interface in hough_line_peaks
      Use new equalize_hist function in rank filter example
      Fix deprecation warnings in matching example
      Fix deprecation warnings
      Update rank filter example
      Fix novice doc test
      Remove print statement
      Fix cython compilation warnings
      Fix custom six issue in scipy 0.14 on python 3.4
      Test invalid input for seed
      Test invalid input for selem
      Add full test coverage for corner_subpix and fix bug
      Test union of differing geometric transforms
      Test 90deg rotation for similarity transform
      Fix docstring of skimage.filter.rank.threshold_percentile
      Test invalid input for unwrap_phase
      Comment special case when local patch around is constant
      Improve description of subpixel corner localization
      Move removal of deprated functionality by one version
      Set version 0.10
      Add 0.10 to docs
      Change to http git protocol
      Update release instructions for building docs
      Fix gh-pages build script
      Increase version to 0.11dev
      Remove deprecated release instruction
      Count number of commits
      Merge pull request #1016 from stefanv/doc/depends
      Merge pull request #1022 from jni/medial_axis
      Merge pull request #1026 from stefanv/bugs/watershed_none_comparison
      Merge pull request #1025 from stefanv/compat/painttool_mpl_NoNorm
      Merge pull request #1027 from stefanv/maint/ignore_generated_notebooks
      Merge pull request #1028 from stefanv/viewer/median_filter
      Merge pull request #1024 from stefanv/bugs/qt_pyside_matplotlib_compat
      Test negative intensities
      Merge pull request #1042 from stefanv/setup_numpy_dependence
      Merge pull request #1086 from jni/thresh-iso
      Add release note instructions
      Describe new release note dev process
      Add release notes for upcoming versions
      Remove contributors for current release notes
      Rename Changes section to API changes
      Add intro to release notes
      Add release notes template
      Fix denoise tests
      Fix isodata-thresholding according to Zachary Pincus
      Return to old function name threshold_isodata
      Account for case when image only contains one unique value
      Fix test cases which are now wrong due to previous binning error
      Improve test coverage
      Fix return value description
      Test missed line for coverage
      Remove legacy CI config
      Merge pull request #1102 from stefanv/bug/freeimage_uint16
      Merge pull request #1105 from pmneila/master
      Merge pull request #1128 from blink1073/fix-travis
      Merge pull request #1109 from blink1073/add-exposure-mask
      Speedup moments_central function
      Speedup moments function
      Make moments functions C functions and reuse centralized moments
      Replace pow() with iterative multiplication
      Merge pull request #1129 from stefanv/log_leftovers
      Merge pull request #1127 from Brittix1023/windowed_histogram
      PEP8 fixes and rearranged images in plot
      Fix spelling
      Replace jet with hot colormap
      Merge pull request #1134 from magcius/fix-star-imports
      Merge pull request #1135 from stefanv/freeimage_big_endian
      Merge pull request #1136 from juliantaylor/signed-char
      Merge pull request #1145 from jni/find-contours-docstring
      Improve desciption of inverse_map and add option to direclty pass coordinates
      Improve inverse_map description
      Make warp function N-D compatible, and add example to doc string
      Remove deprecated parameter of warp
      Highlight fact that warp_coords is only meant for 2(+1)-D images
      Make error message for wrong input dim more explicit
      Add comments to different conditional paths
      Add support for 2(+1)D images and given output_shape for 2D images
      Make row, col ordering clearer in doc string
      Merge pull request #1150 from blink1073/overlay-blit
      Move comments into if scope
      Fix indendation of parameter description
      Remove legacy parameter
      Add option to skip clipping of output, and detect if input float image has negative values
      Extend description of clipping behavior
      Add test for N-D warping
      Use for mgrid for Py26 compatibility
      Merge pull request #1156 from alexeyum/ellipse-symmetry
      Merge pull request #1155 from atemysemicolon/master
      Merge pull request #1159 from blink1073/check-ndims-in-edges
      Add option to combine tform with inverse tform
      Add option to rotate around specific center
      Merge pull request #1169 from stefanv/doc/denoise_opt_param
      Merge pull request #1123 from stefanv/pnpoly
      Add stop_probability to RANSAC
      Add sanity checks for input parameters
      Merge pull request #1183 from jni/warn-single-labels
      Remove extra space
      Fix stop_probability description
      Improve description of stop_probability
      Refer to rank filters from morphology functions
      Add note about supported data types in rank filter replacements
      Clarify that uint16 images are only faster up to a certain bit depth
      Use correct rank filter replacement function
      Add fused type specialization for moments functions
      Use uint8 type for simple moments functions
      Add encoding header
      Add signed edge operators and deprecate unsigned versions
      Add deprecation instructions for old unsigned edge map versions
      Use new edge operator functions in test cases
      Add missing comma
      Remove deprecated absolute edge map description
      Fix rank filter kernels
      Wrap code at 80
      Add tests for all rank filters and reformat code
      Fix bug in bicubic interpolation and extend test cases to higher orders
      Add reference to bicubic interpolation
      Make code PEP8 compliant
      Fix reference syntax
      Fix broken test case, which used black image
      Fix bug in get_pixel3d
      Fix quadratic interpolation
      Use SciPy for second order interpolation, add warnings to buggy functions
      Remove broken test case
      Misc PEP8 fixes
      Merge pull request #1291 from jni/exposure-fix
      Check for invalid upsampling block sizes
      Add test cases for invalid block sizes
      Merge pull request #1295 from stefanv/doc_build_fixes
      Improve code style
      Make the initialization deterministic
      Merge pull request #1298 from pratapvardhan/morph-selem
      Expose seed parameter, fix doc string
      Set seed for test cases
      Merge pull request #1315 from sciunto/ffi
      Clip to min and max range of input image, add missing clip parameter to other functions
      Introduce parameter to keep range of values
      Fix keep_range and add test cases
      Create helper function for image conversion
      Make sure 1x1 images are 2-D
      Remove test for invalid shape
      Rename keep_range to preserve_range, fix test cases
      Remove __init__.py from tests folder, remove unnecessary load of test image
      Fix matching test cases
      Fix novice module with new dtype behavior
      Use in-place operation for clipping
      Add more description to doc string of helper function
      Add missing return value description
      Use typed memoryviews
      Create function to center and normalize image points
      Improve conditioning of homography design matrix
      Improve conditioning of silimarity transform design matrix
      Handle special case of 1 pixel image
      Add new parameter to catch exceptions during RANSAC
      Disable catch of all warnings, return nan params instead
      Fix doc string
      Use relative imports in skimage files
      Resolve incorrect rebase
      Raise Cython dependency to 0.21
      Fix invalid relative import
      Speedup regionprops
      Fix Hessian matrix eigvals
      Fix RANSAC for invalid model estimation and confidence corner case
      Update doctest for new estimator API
      Use np.inf instead of float('inf')
      Fix out-of-bounds access in color LUT for NaN values
      Add decorator to test parallel execution of unit test
      Add unit test for test_parallel decorator
      Do not acquire GIL for warp_fast
      Test parallel execution of warp_fast
      Do not acquire GIL for iradon_sart
      Do not acquire GIL for hough_circle
      Do not acquire GIL for hough_line
      Do not acquire GIL for slic
      Do not acquire GIL for quickshift
      Do not acquire GIL for felzenswalb
      Do not acquire GIL for heap
      Do not acquire GIL for median_filter
      Do not acquire GIL for rank filters
      Do not acquire GIL for ORB
      Do not acquire GIL for corner detectors
      Do not acquire GIL for CENSURE
      Do not acquire GIL for BRIEF
      Do not acquire GIL for texture functions
      Do not acquire GIL for hessian_matrix_det
      Do not acquire GIL for draw functions
      Do not acquire GIL for shared functions
      Replace libc.math.NAN with numpy.math.NAN for MSVC
      Improve documentation of moments functions
      Merge pull request #1583 from grlee77/fix_interp_modes
      Some doc string fixes for recently merged code
      Merge pull request #1684 from juliusbierk/hough-circle-input
      Enforce integral type for label images
      Merge pull request #1667 from odebeir/sprint_euroscipy2015_denoise
      Merge pull request #1670 from odebeir/sprint_example_gamma
      Merge pull request #1575 from blink1073/imageio-plugin
      Merge pull request #1672 from tv3141/add_prewitt_filter_to_example
      Merge pull request #1721 from sciunto/fixurl
      Merge pull request #1758 from JDWarner/hough_test
      Fix region props deprecation problem with label function
      Fix deprecated filter module import
      Fix circular import between regionprops and convex_hull_image
      Remove single author from header, fix typo
      Move import test to top of function and import outside of function
      Remove redundant type conversion
      Fix circular import loop
      Merge pull request #1676 from odebeir/sprint_add_entropy_example
      Improvements to entropy example
      Merge pull request #1795 from kevin-keraudren/ransac-linemodel3D
      Use more common syntax to assign axis variables
      Merge pull request #1764 from glemaitre/geometric_mean
      Fix LineModelND test cases
      Create helper function to compute norm along axis
      Merge pull request #1768 from stefanv/mailmap
      Merge pull request #1815 from stefanv/regionprops_cached_decorator
      Merge pull request #1844 from emmanuelle/fix_euler
      Fix rotation with resizing and custom center
      Fix Gaussian kernels in hessian_matrix
      Fix hessian_matrix doctest
      Fix coordinate extraction
      Some improvements to the doc string and code formatting
      Fix corner orientations unit test
      Raise error for invalid min_distance values
      Fix corner_peaks doctest
      Fix corner_fast doctest
      Fix peak_local_max doctest
      Change default threshold_abs value to minimum image intensity
      Fix BRIEF test cases
      Fix bug in exclude_border option
      Fix matching test cases
      Add relative threshold to corner_peaks
      Merge pull request #1916 from blink1073/warn-stacklevel
      Speedup rgb2gray and return congiguous array
      Make sure rgb2gray output is contiguous
      Fix rgb2grey for alpha channel
      Fix deprecation warning in plot_ransac
      Apply PEP8 to plot_adapt_rgb
      Fix deprecation warning in equalize_adapthist
      Fix deprecation warning in plot_active_contours
      Fix warning in plot_gabors_from_astronaut
      Fix warning in plot_rank_mean
      Fix warning in plot_view_as_blocks
      Fix warning in plot_swirl
      Improve plot_morphology
      Fix unknown Example section warning
      Fix unknown Note section warning
      Fix unknown Properties section warning
      Apply PEP8 to plot_circular_elliptical_hough_transform
      Fix test cases for new deprecation behavior
      Revert contiguous array conversion
      Use kwargs for figure config
      Merge pull request #1921 from JDWarner/image_comparison_prefix
      Merge pull request #1925 from sciunto/test_naming
      Fix some remaining TODO items
      Document API changes
      Remove unit tests for deprecated features
      Merge pull request #1951 from jmetz/update_peak_local_max_exclude_border

John Wiggins (22):
      Raise an error for negative sigma values in guassian_filter.
      Replace a doc test with some regular unit tests.
      Don't use two lines to raise an exception with a short message.
      Use assert_raises instead of @raises.
      Replace cubehelix with the viridis colormap in the RAG drawing example
      Import viridis from skimage.util.colormap.
      Fix some things on Travis
      Revert pip cache changes. Use pip's built-in retry.
      Make sure the pip in the virtualenv is upgraded.
      Remove imread from the optional requirements for 2.7
      Enable ccache on Travis
      Attempt to cache the sphinx-build intermediates on Travis.
      Fix the SPHINXBUILD define for Travis
      Correct the docs build on Travis
      Try again with the sphinx build on Travis.
      Make sure sphinx is installed before building the docs on Travis
      Allow a cache directory to be passed to the docs Makefile
      Put the docs build command on a single line.
      Use the current python version for the python intersphinx reference
      Remove sphinx from the optional dependencies
      FIX: Move a argument check to avoid leaking malloc'd arrays.
      Use np.empty() instead of malloc/free in _denoise_bilateral.

Jonathan Helmus (25):
      ImageCollection now slices like other iterables
      BUG: segmentation fault when unwrapping 3d image
      TST: Unit test to check for 3d unwrap seg fault
      BUG: unwrap_phase on compressed masked array
      Remove end of line whitespace in test_unwrap.py
      unwrap_phase returns masked array when image is masked
      No SegFault when unwrapping a 2D fully masked array
      No SegFault when unwrapping a 3D fully masked array
      image_unwrapped created as ndarray always
      Implement appveyor builds
      TST: Faster test_multi_image unit tests.
      TST: Use smaller gif file in test_pil tests
      TST: Remove no_time_for_that.gif from data
      TST: Use quarter size images to speed up test_censure unit tests
      TST: Reduced size of image in test_blob:test_blob_log for speed
      TST: Smaller data array in 12-bit entropy test for speed
      TST: smaller arrays for bit comparison in test_rank for speed
      BLD: __SKIMAGE_SETUP__ global variable for detecting build
      MAINT: Fixed __check_build setup.py file
      BLD: Updated bento.info with _check_build extension
      Removed __check_build directory
      ENH: Build check in skimage/__init__.py module
      MAINT: remove unchecked import from util.dtype
      MAINT: Refactored imports
      DOC: Simplified the message when a broken skimage build detected

Jonathan J. Helmus (3):
      ImageCollection now supports slicing
      added test for ImageCollection slicing
      variable name changes and small fixes

Josh Warner (89):
      Merge pull request #526 from sciunto/fixtravis
      Merge pull request #545 from odebeir/master
      Merge pull request #551 from tonysyu/pyside
      Merge pull request #558 from chintak/convex_hull
      Merge pull request #571 from ahojnnes/travis-fix
      Merge pull request #588 from sciunto/range
      Merge pull request #575 from tonysyu/feature/viewer-linking
      Merge pull request #634 from josteinbf/arraypad-zeros
      Merge pull request #640 from sciunto/missingdoc
      Merge pull request #637 from sciunto/minorfix
      Merge pull request #656 from josteinbf/doc-makefile
      Merge pull request #603 from TheChymera/master
      Merge pull request #632 from ahojnnes/local-blocks
      Merge pull request #682 from jni/slic-type-hotfix
      Merge pull request #692 from angelatlarge/fix_hog_orient_bins_rounding
      Merge pull request #770 from cgohlke/patch-3
      Merge pull request #789 from ahojnnes/binary
      Merge pull request #787 from cgohlke/patch-2
      Merge pull request #788 from ahojnnes/patch-1
      Merge pull request #847 from ahojnnes/template
      Merge pull request #907 from cgohlke/patch-1
      Merge pull request #926 from sivapvarma/plt-subplots-examples
      Merge pull request #991 from stefanv/apigen_origin_license
      Merge pull request #1019 from stefanv/build/extra_files
      Merge pull request #1069 from jni/nx-depends
      Merge pull request #1112 from pelson/docs_tidy
      Merge pull request #1213 from blink1073/fix-windows-install
      Merge pull request #1221 from stefanv/pr_1126_update
      Merge pull request #1256 from matejak/contributors
      Merge pull request #1258 from blink1073/add-install-requires
      Merge pull request #1289 from ahojnnes/random
      Merge pull request #1344 from sciunto/doc
      Merge pull request #1355 from kevin-keraudren/peak_local_max-NDarrays-issue1354
      Merge pull request #1367 from blink1073/fix-deprecated-decorators
      Merge pull request #1378 from arve0/fix_canny_simple_viewer_example
      Merge pull request #1380 from arve0/fix_Blitmanager.remove_artists
      Merge pull request #874 from emmanuelle/nlm_denoise
      Merge pull request #1391 from ahojnnes/eigvals-fix
      Merge pull request #1413 from ahojnnes/inf
      Merge pull request #1424 from blink1073/contrast-check-function
      Merge pull request #1388 from arve0/move_docstrings_to_init
      Merge pull request #1446 from arve0/readme_typo
      Merge pull request #1457 from blink1073/fix-pkg-resources
      Merge pull request #1468 from stefanv/docs/api_cython_funcs
      Merge pull request #1455 from stefanv/doc/release_notes_and_installation
      Merge pull request #1507 from arve0/pop_bilateral_docstring_typo
      Merge pull request #1513 from joelfrederico/skimage-measure-regionprops-doc
      Merge pull request #1523 from warmspringwinds/ellipse-doc-correction
      Merge pull request #1526 from arve0/installation_instructions
      Merge pull request #1546 from blink1073/window-size
      Merge pull request #1535 from stefanv/doc/installation
      Merge pull request #1550 from jni/ndi
      Merge pull request #1558 from blink1073/optimized-window
      Merge pull request #1555 from sciunto/user_guide-joblib
      Merge pull request #1536 from warmspringwinds/mb-lbp
      Merge pull request #1459 from vighneshbirodkar/seam_carving
      Merge pull request #1562 from warmspringwinds/example_correction
      Merge pull request #1553 from blink1073/tifffile-package
      Merge pull request #595 from TimSC/fasthog
      Merge pull request #1576 from belltailjp/fix-broken-link-to-daisy-implementation
      Merge pull request #1573 from jni/contributing-array-coords
      Merge pull request #1598 from stefanv/bug/line_aa_horiz_vert
      Merge pull request #1565 from blink1073/viewer-ax
      Merge pull request #1607 from blink1073/faster-travis
      Merge pull request #1604 from blink1073/pil-close
      FIX: Require networkx 1.9.1 for the 2.6 build
      FIX/STY: Remove magic and enforce PEP8 in plugins.
      Merge pull request #1638 from ahojnnes/doc-fixes
      Merge pull request #1738 from blink1073/test-3.5
      Merge pull request #1421 from hohlraum/hough_fix
      Merge pull request #1766 from blink1073/fix-setup
      Merge pull request #1797 from ahojnnes/entr-ex
      Merge pull request #1800 from blink1073/fix-docs
      Merge pull request #1799 from ahojnnes/linemodelnd
      Merge pull request #1798 from kevin-keraudren/seamcarving-typos
      Merge pull request #1771 from noahtreb/fix-correct_mesh_orientation
      Guarantee correct mesh orientation from marching cubes
      Merge pull request #1790 from WarrenWeckesser/imread-indexed-png-alpha
      Merge pull request #1877 from OrkoHunter/test_hog_lena_deprecate
      Merge pull request #1902 from OrkoHunter/test_orb_fix_lena
      Merge pull request #1898 from ahojnnes/anntzer-peak-local-max
      Merge pull request #1915 from ahojnnes/corner-fix
      Merge pull request #1897 from grlee77/simple_metrics
      Merge pull request #1899 from OrkoHunter/hog_normalise
      Merge pull request #1817 from blink1073/upgrade-imshow
      Merge pull request #1942 from emmanuelle/trailing_todo
      Merge pull request #1972 from jni/bump-cython
      Merge pull request #1976 from stefanv/mpl_warnings
      Merge pull request #1981 from pradyumnanpk/bilateral_filter-enhancement

Josh Warner (Mac) (111):
      FEAT - combined API from is_local_maximum() into peak_local_max()
      FEAT - Unified peak finder backend
      BUG - Fixed a bug where, when `indices` = False, output was not Boolean.
      BUG - Fixed a bug where the image was not properly masked for each unique label.
      BUG - Now parsing all unique labels != 0 properly, and set threshold_rel=0 in is_local_maximum() wrapper.
      BUG - Dtype conversion to Boolean in a new code path not included in current tests.
      BUG - `rank_order` now imported.
      doc: deprecated string and docstring formatting fixes in `is_local_maximum`
      style: initialize `out` at beginning of function
      doc: fix and clarify `peak_local_max` docstring
      test: add additional tests for 100% coverage of new `peak_local_max`
      fix: set random seed in `test_noisy_square_image`
      workaround: convert `dtype(image)` to `image.astype(dtype)`
      FIX use specific structuring element in Canny
      fix to preserve input image shape upon executing `random_walker`
      PEP8 fixes; remove unneeded pyamg import; add `data` shape check
      Use `force_copy` argument of `img_as_float`
      fix: remove unnecessary forced copy
      add assertion checks to ensure `data` shape is unchanged
      temporarily use Cython 0.18 to restore normal Travis builds
      Fix Cython 0.19 build error due to global wraparound=False
      DOCFIX: "\" in `label` docstring now escaped
      FEAT: Generalized n-dimensional array padding
      FIX: prepend `np.testing.` to assert_* statements
      Use memoryviews in skimage.transform
      FEAT: Automatically switch between RGB / grayscale lineprofile
      DOC: Add viewer example for RGB line profile
      FIX: refactor code, fix linewidth calculation
      FIX: lineprofile no longer flips on left half and cardinals work
      FIX: remove blit, fix 0-length error on grayscale images
      FIX: only reset axes when gray <-> rgb, not every update
      FIX: No longer reset for each update
      FEAT: generator to add various types of random noise to images
      ENH: Improve random_noise with better param names & docfixes
      FEAT: Full 3d anisotropic implementation of marching cubes
      import `marching_cubes` and `mesh_surface_area`
      add _marching_cubes.pyx to list of Cython modules to build
      fix import of _marching_cubes
      fix: bug in second `_append_tris` call in case 125
      rename files to conform with naming conventions
      fix import error from `_marching_cubes_cy`
      fix error in triangulation of case 120
      FIX: dimension ordering now correct (x, y, z)
      FIX: volume now directly cast, docfixes, C-contiguous ordering
      FIX: use memoryviews, docfix, and change tri_list to face_list
      FIX: incorrect name for `unpack_unique_verts`, revert axis ordering
      DOCFIX: clarify docstrings and output vertex dimension ordering
      FEAT: add tests for marching cubes and mesh surface area
      DOC: Minor documentation formatting fixes in marching cubes
      MAINT: Refactor ellipsoid generator into skimage.draw
      DOC: Add concise marching cubes example to gallery using Matplotlib
      TEST: Add test suite for draw3d (ellipsoid/ellipsoid_stats)
      FIX: Add _marching_cubes_cy.pyx to bento.info
      FIX: Errors in unit tests
      STYLE: Use `np.array` instead of `np.r_` and import Poly3DCollection
      Credit myself for Marching Cubes
      DOC: Further improvements to naming conventions.
      FIX: Fix and improve Poisson random noise generator
      ENH: More concise next-power-of-2 calculation
      ENH: Add optional `clip` kwarg, docs, & handling of signed inputs.
      ENH: Tighten tests, all noise types now support signed I/O
      FIX: Better handling of skimage.data.camera for Python3 compatibility
      FEAT: Add 'localvar' mode to random_noise
      REBASE: Resolve first conflict
      ENH: `spacing` kwarg for random_walker and improved tests
      FIX: roll back incorrect testing change of gradient order
      DOC: Add removal of `depth` from random_walker in 0.10
      DOC: Fix minor typo in docstring
      DOC: Change sampling kwarg name to spacing in marching_cubes
      DOC: Change sampling to spacing in skimage.draw.ellipsoid and tests
      DOC: Fix capitalization in ellipsoid docstring.
      FEAT: Add function to correct mesh face orientations
      Add correct_mesh_orientation to skimage.measure imports
      DOC: correct misspelling in comment
      TST: Add unit test for mesh orientation correction.
      FIX: Slight speed decrease for backwards compatibility
      FIX: correct other instance of np.linalg.norm, clarify comments
      DOC: Clarify UMFPACK warning; use if statements instead of assert
      FIX: Shortcut output or catch trivial cases in random_walker
      TST: Add unit test for length-2 spacing with rank-2 input
      STY: More elegant and maintainable style per code review
      FEAT: Shared function for safe int casting
      Remove check for non-numeric values in spacing.
      Convert name to _safe_as_int, output np.int64, expand Examples
      FIX: Remove unnecessary .transpose introduced in PR #984
      failUnless -> assertTrue
      FIX: Address non-int DeprecationWarning from NumPy
      FIX: Non-integer indexing deprecation warnings
      Fix documentation, handle internal variable more efficiently.
      Remove incorrect np.round statement (directly cast `m`)
      Use safe_as_int for input checking.
      Merge branch 'master' of git://github.com/scikit-image/scikit-image into int_idx_patch
      Remove unnecessary double import.
      FIX: Use //= instead of /= to fix deprecation warnings in adapthist
      ENH: Simplify marching_cubes for maintainability.
      ENH: Fully cover `pad` and backports from NumPy 0.10dev
      FIX: Informative IOError if package installed w/mixed permissions.
      FIX: Ignore exception if attempted deletion fails (fixes #817)
      DOC: Fix Sphinx rendering of dev process docs
      FIX: Use direct, secure link for pip wheelhouse.
      FIX: typo in local version of tifffile _replace_by
      DOC/STY: Fix whitespace, harmonize version with PyPi 0.4.1
      ENH: `make clean` now removes compiled .c and .pyd files.
      Remove only .c files associated with .pyx source.
      TSTFIX: Missing assert statement and PEP8
      DOC: Link to GitHub source in documentation
      DOC: Fix intersphinx mapping w/fallbacks
      TST/STY: Hough tf regression test & Cython wrappers
      TSTFIX: Fix imports in hough_ellipse doctest
      DOC: Correct docstring to correctly reflect defaults
      STY: Fix formatting and use tight_layout in doc examples.

Joshua Warner (4):
      FIX: Point Windows batch file at source subdir.
      STY: Image comparison functions now share common prefix
      FIX: Pass arguments from deprecated function, update examples.
      DOC: Note removal of structural_similarity in TODO.txt

Jostein Bø Fløystad (152):
      Add reconstruction circle option to transform.radon.
      Tests for reconstruction circle option for transform.radon.
      Add reconstruction circle option to transform.iradon.
      Tests for reconstruction circle mode in transform.iradon.
      Correct documentation and string formatting.
      PEP8 style changes for transform.radon_transform and its tests
      Style improvements in docstrings in transform.radon_transform
      Update contributors
      Reduce code duplication in test_radon_transform
      PEP8 fixes for test_radon_transform
      Radon transform example: Correct axis labels
      Radon transform example: Remove semicolons.
      transform.radon: Add test to verify the projection center.
      transform.radon: Add testcases for rectangular input arrays.
      transform.iradon: Add tests for center of projection.
      transform.radon: Remove unneccesary matrix inverse.
      transform.radon: Use correct padding for rectangular images.
      transform.radon: Consistent definition of center of array.
      radon tests: Refactor test for circle reconstructions.
      transform.iradon: Redefine slice and projection center.
      test_radon_transform: Add helper functions.
      transform.iradon: Refactoring for shorter functions.
      test_radon_transform: Test sinogram conversions.
      test_radon_transform: Refactor and improve test_radon_iradon.
      test_radon_transform: debug option for test_iradon_minimal.
      radon_transform: Stylistic changes.
      test_radon_transform: Refactor tests and make them stricter.
      test_radon_transform: Change test criterion for iradon_minimal.
      test_radon_transform: Change test criterion for sinogram_circle.
      transform.radon: Robust determination of center of projection.
      transform.iradon: Correct determination of center of projection.
      test_radon_transform: Clean up imports.
      test_radon_transform: Style fixes, PEP8.
      Add SART tomography reconstruction to radon_transform.
      Make iradon_sart visible in the transform package.
      Add tests for transform.iradon_sart.
      Implement projection shifts.
      Tests for the projection_shifts functionality of iradon_sart.
      Fix docstrings for transform.iradon_sart with subroutines.
      transform.iradon_sart: Clean up code for ordering projections.
      iradon_sart: Add wikipedia reference to Kaczmarz method.
      iradon_sart: Clarify how constants are chosen.
      Tests for iradon_sart: Robust path handling.
      iradon_sart: Clean up cython code to minimize python calls.
      iradon_sart: Improve argument checking.
      iradon_sart's cython exts: Use typed memoryviews, not ndarrays.
      iradon_sart: Reduce code duplication in interpolation.
      iradon_sart: Remove needless memoryview/ndarray conversion.
      iradon_sart: Add Kaczmarz reference and reformat citations.
      radon_transform: Declare encoding.
      iradon_sart: Order angles using a golden ratio approach.
      iradon_sart: Add test for order_angles_golden_ratio.
      iradon_sart: Also test accuracy with a missing wedge.
      iradon_sart: Style fixes.
      test_radon_transform: Style fixes.
      Rewrite Radon example; include SART.
      iradon_sart: Test clip functionality and add a test for it.
      radon example: Display plots when run from command line.
      Improve the text in and add references to the Radon example.
      Radon example: Style fixes.
      Radon example: print RMS errors of the reconstructions.
      Radon example: Declare UTF-8 encoding.
      Radon tests: Check the correct error metric.
      iradon_sart: redefine projection center.
      iradon_sart: fix comment spelling.
      iradon_sart: flip signs to reflect changes to radon.
      Add SART to contributors.
      arraypad: allow padding with zero entries (i.e. no padding).
      radon: Use util.pad for array padding.
      iradon: Use util.pad for array padding.
      iradon: Reduce code duplication.
      radon: Use numpy.interp/scipy.interpolate.
      iradon: Only allow interpolation methods working well.
      iradon: Add test for cubic interpolation.
      iradon: Cleanup by locating related code in one place.
      iradon: use util.pad for sinogram padding.
      iradon_sart: Format docs correctly.
      iradon_sart: Reformat references.
      iradon_sart: Fix docstring typo.
      examples: Avoid python3-specific syntax.
      Radon transform: Redefine projection center for sinograms.
      Radon transform: Document rotation axis location.
      radon: Reduce duplication; simplifications.
      radon: Correct docstring of order_angles_golden_ratio.
      Radon transform: Include boundary in reconstruction circle.
      Radon transform: PEP8 fixes.
      iradon: Clean up filter code.
      iradon: Do not suppress 0 frequency for filter=None.
      docs: Makefile respects variables from environment.
      doc: Allow specifying python binary.
      Bug: Fix label sequence in regionprops.
      unwrap: Move files to skimage directory tree.
      unwrap: Delete old setup script.
      unwrap: Remove executable bit.
      unwrap: Rename cython files to skimage conventions.
      unwrap: Add setup script to build extensions.
      unwrap: Use new imports in python module.
      Make unwrap visible in the exposure package.
      unwrap: refactor tests.
      unwrap: Whitespace fixes.
      unwrap: Move imports to module level.
      unwrap: Fix whitespace in c code.
      unwrap: Add unfinished docstrings.
      unwrap: Refactor wrap_around argument.
      unwrap: Refactor to skimage-style variable names.
      unwrap: Add docstring.
      Add example for phase unwrapping.
      unwrap: Use relative imports
      unwrap: Correct use of relative imports.
      unwrap: Format references correctly.
      unwrap: Remove plotting code.
      unwrap: Add citation for 3D.
      unwrap: reformat references
      unwrap: Remove unused imports
      unwrap: Simplify handling of masked arrays.
      unwrap: Test wrap_around functionality.
      unwrap: Refactor and fix wrap around functionality.
      unwrap: Whitespace fixes and comment removal.
      unwrap: rename cython functions.
      unwrap: PEP8 fixes for cython extensions.
      unwrap: rename variables in cython extensions
      unwrap: Update bento.info.
      unwrap: Python 3 compatibility.
      unwrap: PEP8 fixes.
      unwrap: Reduce duplication in tests.
      unwrap: Reduce code duplication.
      unwrap: Make sure test data wraps along all dimensions.
      unwrap: Delete "example" from the test file.
      unwrap: Add test for unwrapping masked images.
      unwrap: Simplify example.
      unwrap: PEP8 fixes.
      unwrap: Small style-like fixes.
      unwrap: Rename function unwrap() -> unwrap_phase().
      unwrap: Warn about singleton dimensions in 3D arrays.
      unwrap: Add docstring example.
      unwrap: Rename C files.
      unwrap: Add naive 1D unwrapper.
      unwrap: Use constants from math.h in C code.
      unwrap: Convert 2D C code from float to double.
      unwrap: Use UNIX line endings.
      unwrap: Change 2D wrapper code from float to double.
      unwrap: Change 3D C code from float to double.
      unwrap: Change 3D wrapper code from float to double.
      unwrap: Change 1D unwrapper from float to double.
      unwrap: Simplify code after completing float->double.
      unwrap example: cleanup.
      unwrap example: Mention dimensionalities available.
      unwrap example: Include masking and wrap around.
      unwrap tests: PEP8 fixes.
      unwrap: Do not include debug symbols in extension builds.
      Less verbose entry in CONTRIBUTORS.
      Add contributors for phase unwrapping.

Juan Nunez-Iglesias (451):
      Add function 'concatenate' to ImageCollection
      Add test for ImageCollection.concatenate
      Add concatenate() support for MultiImage
      Sort files from a global pattern alphanumerically
      Add test coverage for alphabetic sort
      Simplify alphanumeric_key logic
      Add functions to join and relabel segmentations
      Add testing functions for join and relabel
      Add join_segmentations example to the gallery
      Rename ar_relabeled to relabeled for simplicity
      Tweak join_segmentations example for clarity
      Rename `ar` to `label_field` for clarity
      Allow grayscale input images to SLIC segmentation
      Add funct to remove small connected compononents
      Add examples to docstring
      Add tests for remove_small_connected_components
      Modify example to use remove_small_connected_components
      Improve docstring for remove_small_connected_components
      Rename function to remove_small_objects
      Add raise and labeled image tests
      Fix missed rename in example
      Add specific error messages and exceptions
      Revert remove_small_objects to raise TypeError
      Make `exclude_borders` option in peak_local_max nD
      Use range instead of xrange
      Make color functions used by SLIC 3D-aware
      Make more colorconv functions 3D aware
      Add initial 3D modifications (not working)
      Add function to calculate regularly-spaced grid in nD
      Modify SLIC to allow uneven step sizes
      Add more descriptive error message to _prepare_colorarray
      Separate inner loop of SLIC computation
      Bug fixes: concatenation order and shape assignments
      Bug fix: don't add singleton dimension to 3D gray images
      Bug fix: remove unnecessary __get__ in test_slic
      Add 3D slic tests (gray not working yet)
      bug fix: 3d test working
      Improve PEP8 and Python 3 compliance
      PEP8
      Always convert image to float in [0, 1]
      Rename slic.py to avoid name conflicts
      Use educated guesses for presence of color channels in SLIC
      Add function to guesstimate number of spatial dimensions
      Revert change to is_rgb
      Bug fix: add z dimension regardless of multichannel
      Bug fix: spatial vs image dimension comparison reversed
      Make means array contiguous for numpy <= 1.6.1
      Add UTF8 coding declaration to slic_superpixels.py
      Suppress warning of ambiguous array dim in test
      Initial attempt at updating SLIC for memoryviews
      Fix build errors with memoryviews
      Improve _slic.pyx doc, bug fixes, debug print
      Remove diagnostic print
      Bug fix: correctly initialize distance in slic
      Speed up initialising 'distance' in SLIC
      Fix automerge bug
      Bug fix: map returns iterator in Py3k
      Remove old commented-out code
      Improve doc for regular_grid.py
      Add tests for 'regular_grid()'
      Rename regular_grid.py
      Add test coverage for guess_spatial_dimensions()
      Complete test_regular_grid coverage
      Fix typo using assert_raises in test_colorconv.py
      Fix ratio scaling in SLIC
      Merge pull request #662 from stefanv/bento_installation
      Remove unused 'ratio' argument from _slic_cython
      Precision issues appear to have vanished
      Replace 'ratio' kwarg with 'compactness' in SLIC
      Do not use 'ratio' in test_slic.py either
      Add unique_rows function
      Add test for unique_rows
      Add unique_rows to util/__init__.py
      Remove repeated coordinates before computing convex hull
      Unique rows after getting the pixel corners
      Move code comment to appropriate line
      Add (previously) pathological qhull test case
      Add test to unique_rows for discontiguous arrays
      Ensure array input to unique_rows is contiguous
      Use Py_ssize_t instead of int in _slic_cython
      Change 'long' to 'int' in Cython docs
      Merge pull request #689 from mkcor/coffeepic-credits
      Merge pull request #665 from mrterry/deltae
      Add note describing array copy if discontiguous
      Add docstring note explaining coord use case
      Add PyAMG to the optional dependencies list
      Update PyAMG URL (no longer in Google Code)
      Merge pull request #712 from emmanuelle/gaussian_filter
      Fix errors in relabel_from_one docs; PEP8
      Fix type hierarchy check in remove_small_objects
      Fix image dimension sanitizing at function start
      Update SLIC docstring to remove deprecated example
      Ignore `convert2lab` keyword if not multichannel
      Add support for list sigma input in SLIC
      Add spacing support for new, speeded-up SLIC
      Add `spacing` descriptions; use np.double
      Bug fix: typo: wrote spacing instead of sigma
      Add note about (z, y, x) order in slic cython doc
      Update snippets in notes to use double backticks
      Merge pull request #719 from ahojnnes/slic
      Merge pull request #740 from ahojnnes/warp-fix
      Merge pull request #748 from ahojnnes/regionprops-dtype
      Add relabel_sequential, deprecate relabel_from_one
      Add relabel_sequential to __init__ import
      Add new tests for relabel_sequential
      Remove deprecated use of relabel_from_one
      Rename deprecated relabel_from_one in docstring example
      Only raise umfpack warning if cg mode is used
      Monkey-patch UmfpackContext __del__
      Remove unused imports in random_walker source
      Add comment and links explaining Umfpack import
      Deprecate default mode 'bf' in random_walker
      Add change in random_walker default mode to TODO
      Update SLIC in example to most recent interface
      Merge pull request #761 from kemaleren/optimized_colorlabel
      Fix join_segmentations example using img_as_float
      Merge pull request #774 from ahojnnes/shape-plot
      Merge pull request #769 from tonysyu/consistent-color-labels
      Merge pull request #753 from stefanv/pr_plots
      Merge pull request #779 from ahojnnes/log-depr
      Merge pull request #791 from stefanv/regionprops_as_dict
      Merge pull request #794 from ahojnnes/lbp-memview
      Modify mpl_image_to_rgba to allow RGB input image
      Add plugin return values for ImageViewer
      Allow OverlayPlugin to return the current overlay
      Incorporate @tonysyu's suggestions re:docstrings
      Add output for color histogram plugin
      Merge pull request #839 from rtack/match_template_DOC
      Explicity cast start, step to int in regular_grid
      Merge pull request #856 from ahojnnes/rank-sum
      Merge pull request #862 from tonysyu/refactor/imread_collection
      Add profile_line to measure package
      Use the new profile_line function from `measure`
      Change profile line API to use src and dst points
      Update lineprofile plugin to use new API
      rename x-y to col-row to be less ambiguous
      Merge pull request #864 from romaniukm/slico-rebase
      Update _calc_vert for simplicity
      Return 1D array for single channel images
      Allow multichannel images with >3 channels
      Fix doctest string and output shape
      Reparameterize profile_line to avoid the _calc_vert hack
      Update interactive lineprofile tool to new API
      Simplify parameters even more
      Add 'order' kwarg for interpolation in profile_line()
      Polish coordinate system
      Add numerous (14) tests for profile_line()
      Remove unnecessary if clause
      Change default interpolation order to 1
      Output line drawing and scan data from plugin
      Update docstring according to @stefanv's comments
      Merge pull request #873 from jmetz/convex_hull_image_qhull_precision_fix
      Merge pull request #875 from blink1073/lineprofile-overlay
      Merge pull request #886 from stefanv/bug/pil-fromstring_to_bytes
      Merge pull request #846 from stefanv/coverage-oneversiononly
      Merge pull request #897 from tonysyu/named-rescale_intensity-params
      Merge pull request #913 from jaimefrio/quicker-sort
      Merge pull request #932 from sciunto/weird
      Merge pull request #934 from sciunto/coverage2
      Merge pull request #963 from tonysyu/feature/tinting-example
      Merge pull request #964 from ahojnnes/submods
      Merge pull request #968 from sciunto/req
      Remove deprecated `ratio` keyword in SLIC
      Remove deprecated sigma default warning in SLIC
      Remove SLIC deprecations for 0.10 from TODO.txt
      Remove 'ratio' kwarg from SLIC example
      Remove is_gray and is_rgb
      Remove finished entry from TODO.txt
      Remove all deprecated rank filter functions
      Remove removing deprecated rank functions in to-do
      Remove deprecated parameter in LineProfile
      Remove another finished line from todo
      Change default mode cascade in random_walker seg
      Remove deprecated depth parameter in random_walker
      Remove `depth` from random_walker tests
      Remove deprecated bullet point about deprecations
      Merge pull request #972 from stefanv/test/travis_warnings
      Remove deprecated function from applications
      Fix `plot_segmentations` example text for SLIC
      Update viewer example to use plugin outputs
      Update lineprofile examples to use returned output
      Update plugin class doc example to use output
      Update Canny examples
      Merge pull request #1001 from JDWarner/rw_transpose
      Update variable name for RGB line profile
      Merge pull request #1002 from JDWarner/asserttrue
      Update description of viewer output in docs
      Very minor text change
      Merge pull request #1006 from stefanv/doc/regionprop_out
      Merge pull request #1003 from JDWarner/join_dep_warning
      Clarify `mask` argument in `medial_axis` docstring
      Update parameter list in docstring for consistency
      Update label2rgb doc
      Split label2rgb into two related functions
      Add background label handling for label2rgb_avg
      Add docstring for _label2rgb_avg
      Add default values for background in _label2rgb_avg
      Add docstring for _label2rgb_overlay
      Add tests for label2rgb avg mode
      Merge pull request #1034 from ahojnnes/label-coverage
      Merge pull request #1031 from vighneshbirodkar/rag
      Merge pull request #1052 from charris/numpy-1.9.0-fix
      Merge pull request #1058 from nlsn/default-selem
      Document NetworkX dependency in DEPENDS.txt
      Merge pull request #1068 from vighneshbirodkar/rag_fix
      Merge pull request #1063 from nlsn/3d-fallback
      Merge pull request #1072 from blink1073/add-viewer-tests
      Merge pull request #1073 from tonysyu/feature/adapt_rgb
      Merge pull request #1054 from blink1073/add_test_seeds
      Bug fix: nbins can be >256 in threshold_isodata
      Fix incorrect test function signature
      Merge pull request #1084 from blink1073/miniconda-travis
      Merge pull request #1093 from tonysyu/clarify-release-procedure
      Merge pull request #1090 from ahojnnes/isodata-fixes
      Merge pull request #1083 from adamfeuer/import_export_pil_image_objects
      Merge pull request #1099 from blink1073/build_update_followup
      Merge pull request #1080 from vighneshbirodkar/ncut
      Merge pull request #1106 from vighneshbirodkar/ncut_fix
      Merge pull request #1087 from vighneshbirodkar/rag_draw
      Merge pull request #1096 from ahojnnes/regionprops-speedup
      Merge pull request #1137 from sciunto/doc
      Update find_contours docstring to use r/c notation
      Merge pull request #1151 from blink1073/default_fallback_doc
      Merge pull request #1146 from blink1073/master
      Merge pull request #1142 from ahojnnes/warp-coords
      Merge pull request #1163 from blink1073/implement_assert_nD_array
      Run restoration test suite when run as main
      Add test for preserving image shape in deconv
      Add shape fix for deconvolution functions
      Fix small errors in the documentation
      Fix typos and grammatical errors in uft.py docs
      Warn users of `remove_small_objects` about type
      Test type warning from remove_small_objects
      Merge pull request #1191 from ahojnnes/morph-ref
      Add Travis and Coveralls badges to README
      Merge pull request #1188 from vighneshbirodkar/rag_merge
      Merge pull request #1214 from blink1073/remove-null-plugin
      Remove contiguous conversion for out array
      Add tests for discontiguous out arrays in grey morpho
      Raise meaningful error in reconstruct w bad method
      Test correct error for wrong method in reconstruct
      Merge pull request #1078 from tonysyu/example/adapt_rgb
      Merge pull request #1230 from emmanuelle/clahe
      Protect exposure.histogram from integer overflow
      Merge pull request #1229 from almarklein/patch-1
      Merge pull request #1235 from vighneshbirodkar/blob_fix
      Merge pull request #1239 from ahojnnes/moments-templ
      Merge pull request #1243 from emmanuelle/filters_rank
      Merge pull request #1255 from dedalusj/regionprops-docs
      Merge pull request #1261 from emmanuelle/doctest
      Merge pull request #1269 from blink1073/test-all-modules
      Merge pull request #1272 from blink1073/fix-tifffile-warning
      Merge pull request #1290 from pratapvardhan/morph-selem
      Use integer bins when histeq'ing int images
      Merge pull request #1301 from stefanv/gitter_badge
      Merge pull request #1304 from pratapvardhan/misc-doc
      Merge pull request #1302 from ahojnnes/warp
      Merge pull request #1321 from ahojnnes/hough
      Merge pull request #1313 from blink1073/suppress-test-warnings
      Merge pull request #1325 from blink1073/remove-label-shim
      Initial work on array dimension naming guide
      Restore NumPy from Numpy, grudgingly
      Replace 'level' and 'plane' with 'frame'
      Define 'prefetching' for users
      Finish code snippet for timings
      Fix and add output for timing calls
      Add paragraph about data locality
      Label transpose operation for clarity
      Prefer row/col to r/c when talking about circles
      Clarify boolean masks on color images
      Give 3D image examples
      Space comment and code in plot directive call
      Add default bg_color to label2rgb
      Add test to ensure default bg_color in label2rgb
      Add docstring to matplotlib imshow plugin
      Overhaul matplotlib imshow plugin
      Improve mpl imshow plugin docstring
      Fix incorrect call to colorbar
      Replace Axes with AxesImage where relevant
      Add tests for io.imshow mpl plugin
      Add tests for nonstandard images
      Rename ax to ax_im in mpl imshow plugin
      Use different colormaps depending on input image
      Overhaul imshow to clarify flow
      Add test for signed images
      Don't create new plot automatically
      Expand the docstring for imshow
      Create new figure for each imshow test
      Don't create superfluous ip variable
      Add reference to SLIC0 in SLIC docstring
      Update treatment of convert2lab in slic
      Merge pull request #1340 from blink1073/fix-filter-pkg-shim
      Add underscore for ReST reference
      Wrap scipy.ndimage grey morphology functions
      Wrap scipy.ndimage binary morphology functions
      Workaround for strange ndimage selem inversion
      Update morphology dtypes tests
      Minor update to binary morphology docstrings
      Add crop function, inverse to pad
      Add test functions for crop
      Fix incorrect crop tests
      Fix automatic copy even when copy=False in crop
      Import crop into util package namespace
      Fix and test 0-end bug in crop
      Fix morpho open/close edge bugs by padding image
      Make find_boundaries symmetric
      Fix order kwarg for some versions of numpy.copy
      Update binary morphology tests, which no longer warn
      Update find_boundaries with new modes
      Fix 'outer' mode of find_boundaries
      Remove unused deprecation warning
      Add correct subpixel boundary estimation.
      Set up correct type for find_boundaries test
      Make subpixel find_boundary doctest more readable
      Allow different modes in mark_boundaries
      Prevent unnecessary array copy for RGB images
      Allow custom background_label in mark_boundaries
      Don't outline boundaries by default
      Simplify code for coloring boundaries
      Test subpixel mark_boundaries
      Specify mode for mark_boundary tests
      Fix incorrect call to mark_boundaries in test
      Clarify and fix docstrings
      Replace em-dash with regular dash in comment
      Expand explanation of _invert_selem
      Don't use backticks to refer to functions
      Remove unused cmorph extension
      Fix a few typos
      Add doctest skip directives to timing calls
      Add table summarising image types
      Add note on 5D images
      Expand on skimage support for 3D processing
      Be more specific about cat image dimensions
      Specify standard Cartesian coordinates
      Test symmetric selems for rank/morpho comparison
      Add a paragraph or two about preserving range
      Merge pull request #1351 from ahojnnes/reg-prop
      Fix line wrapping
      Merge pull request #1100 from vighneshbirodkar/ha
      Merge pull request #1359 from blink1073/fix-novice-from-size
      Merge pull request #1360 from blink1073/fix-cmyk-jpg
      Replace frame with plane
      Update discussion on time-varying images
      Move RAG files to skimage.future
      Move package to skimage.future.graph
      Update bento file to moved RAG code
      Update plot_rag examples to use skimage.future
      Update plot_ncut example to use skimage.future
      Add package docstring for skimage.future
      Fix setup.py cascade
      Update RAG doctest to use future
      Fix even more tests still importing skimage.graph
      Import graph into future parent package
      Add concrete usage examples for preserve_range
      Merge pull request #1374 from blink1073/fix-fallback-test
      Merge pull request #1375 from tonysyu/fix/pil-compatibility
      Merge pull request #1385 from stefanv/README_deps_update
      Merge pull request #1412 from blink1073/contrast-check-function
      Clarify 'low-contrast' in docstring
      Update See Also to References
      Merge pull request #1440 from blink1073/fix-imread
      Allow keyword arguments to imsave for PIL plugin
      Add temporary_file context manager
      Add missing comment '#' markers for doctest skip
      Merge pull request #1502 from blink1073/test-mulitpage-rgb-tiff
      Merge pull request #1483 from ozak/ravel-not-flatten
      Merge pull request #1530 from LeeKamentsky/master
      Merge pull request #1537 from kevin-keraudren/ellipse-round-radius
      Merge pull request #1540 from vighneshbirodkar/ncut_fix
      Remove unused variable in example
      Harmonize all ndimage usage across the library
      Specify ndimage as ndi import convention
      Split scipy.ndimage & spatial into different lines
      Merge pull request #1568 from sciunto/threshold_li_fix
      Merge pull request #1524 from bennlich/faster-skimage-draw-ellipse
      Point contributors to coordinate conventions
      Ask contributors to wrap Cython functions in Python
      Use Sphinx links to reference from contributing guide
      Add blank line after ref to keep Sphinx happy ;)
      Merge pull request #1572 from csachs/cython_skeletonize
      Merge pull request #1547 from blink1073/adapthist-window
      Merge pull request #1609 from blink1073/quiet-travis
      Merge pull request #1599 from blink1073/viridis
      Merge pull request #1636 from ahojnnes/moments
      Merge pull request #1657 from robintw/doc-readme-typo
      Merge pull request #1658 from Guanaco77/master
      Merge pull request #1666 from rougier/convex_hull_example
      Merge pull request #1673 from oew1v07/watershed_markers
      Merge pull request #1671 from jwiggins/fix-cubehelix-ref
      Merge pull request #1674 from robintw/canny-percentile-thresholds
      Merge pull request #1725 from sciunto/unclosedfile
      Merge pull request #1744 from blink1073/always-use-tiff
      Refine docstring for 'compactness' kwarg in SLIC
      Replace semicolon with full stop
      Merge pull request #1762 from blink1073/fix-clip-histogram
      Merge pull request #1763 from glemaitre/laplace-edge-detector-wrapper
      Merge pull request #1785 from pratapvardhan/colordir
      Merge pull request #1689 from oew1v07/remove_small_holes
      Merge pull request #1787 from ahojnnes/rpropsfix
      Merge pull request #1794 from kshitij10496/boundaries-bool
      Merge pull request #1449 from stefanv/clear_border_non_binary
      Merge pull request #1770 from stefanv/list_regionprops
      Merge pull request #953 from syedTabish/Issue949
      Merge pull request #1818 from blink1073/improve_downcast
      Merge pull request #1820 from blink1073/deprecate-2.6
      Merge pull request #1819 from blink1073/doc-building-docs
      Update docstring comments for integrate function
      Add deprecation warning for old syntax for integrate()
      Add note in TODO for deprecated use of integrate()
      Update incorrect version number in docstring
      Allow passing individual coordinate tuples to integrate
      Test individual coordinates for integrate in doctest
      Restore vectorised integrate deprecated behaviour
      Merge pull request #1825 from stefanv/travis_pyamg
      Merge pull request #1792 from scottsievert/fftconvolve
      NetworkX is required so remove optional import
      Simplify reference to generate_binary_structure
      Make RAG generic, requiring only label_image
      Add tests for generic RAG construction
      Fix typo in variable name
      Use a strided 1-element array as dummy filter out
      Move max_id logic to top of RAG constructor
      Bug: not all edges found by asymmetric footprint
      Merge pull request #1828 from stefanv/radon_circle_warn
      Update outdated docstring of _add_edge_filter
      Merge pull request #1830 from scottsievert/fftconvolve
      Merge pull request #1841 from blink1073/fix-matplotlib-2.6
      Specify 2D images in denoise bilateral docstring
      Replace individual errors by 2D decorator
      Make area a cached property; use np.sum
      Merge pull request #1505 from emmanuelle/regionprops3d
      Make mean boundary RAG testing more robust
      Use function to update graph edges from CSR
      Hold pixel count in mean boundary RAG edges
      Merge pull request #1495 from vighneshbirodkar/edge_rag
      Merge pull request #1876 from OrkoHunter/threshold_otsu_grayscale
      Merge pull request #1804 from soupault/inpainting
      Merge pull request #1839 from grlee77/nd_denoise_tv
      Merge pull request #1904 from OrkoHunter/clarify_generic_filter_rag
      Merge pull request #1903 from OrkoHunter/threshold_otsu_min_color
      Merge pull request #1907 from OrkoHunter/threshold_adaptive_block_size
      Merge pull request #1913 from stefanv/contribute_2plus
      Merge pull request #1827 from stefanv/daisy_aa_draw
      Merge pull request #1327 from endolith/patch-1
      Merge pull request #1936 from ahojnnes/depr
      Merge pull request #1944 from sciunto/temp_file
      Merge pull request #1923 from ev-br/skel3d
      Merge pull request #1956 from emmanuelle/label_zero
      Merge pull request #1960 from sdsingh/profile_line
      Bump Cython version to 0.23 in DEPENDS.txt
      Merge pull request #1982 from stefanv/orb_no_of_keypoints

Julian Taylor (2):
      improve performance of equalize example
      BUG: add signed flag to chars

Julien Coste (9):
      Move canny from filter to feature
      remove auto added __author__ in __init__.py file
      fix wrong version number
      update api_changes
      fix import inside documentation and update TODO
      fix import in filter.__init__
      move import to avoid infinte recursion (filter import peak that import feature that import filter)
      Canny is now deprecated and will be in api_changes in v0.13
      hack to avoid circular import when import canny

Julius Bier Kirekgaard (19):
      Example for hough_circle
      Added whitespace to comply with pep8
      Shorter hough_circle example
      Fixed forgotten output for docstring example
      hough_circle radius accepts scalars and lists
      test and docstring update for hough_circle
      assert equal scalar, seqeunce output of hough circle
      Added pep8 spaces
      Added active contour model
      pep8 compliance
      Active contour example added
      pep8 and other small changes
      spelling corrections and misc.
      change test names
      Active contour updates. Now works with scipy<0.14
      pep8 and py3 compliance. more comments
      py3 fixes and skip test for old scipy.
      Fixed missing scipy import
      Skip doctest

Julius Bier Kirkegaard (3):
      Fixed doctest problem
      Merge pull request #1 from emmanuelle/active-contour
      Merge pull request #2 from emmanuelle/active-contour-model

K.-Michael Aye (4):
      Fixed 2 typos and 2 wrong argument types in the docs.
      removing some unused selem declarations
      removing obsolete (or not found?) link
      removing unnecessary loop counter

Kemal Eren (3):
      ENH: optimized label2rgb()
      fixed error in label2rgb() when label is a binary array
      handle case when there are no labels

Kevin Keraudren (14):
      the output of hough_ellipse needs to be rounded before being truncated
      code simplification
      Adding LineModel3D for RANSAC, unit test and example.
      reformatting code to wrap at 80 characters per line
      fixed typo and added blank line
      Merged LineModel3D with LineModel in order to form a unified model.
      we now have the old LineModel with params (dist, theta), and the new LineModelND with params (origin, direction).
      cleaned tests and documentation, added deprecation note in doc for LineModel.
      added deprecation warning for LineModel and added note in TODO.txt
      documentation updates; added "3D" to example title
      fixed typos in seam carving example.
      text improvement in seam carving example.
      small text improvement in seam carving example
      text improvement in seam carving example, added some newlines between plt.imshow() and """PLOT2RST"""

Kevin Murray (5):
      [skimg.io._plugins.freeimage_plugin] fix segfault
      [freeimage tests] Add a test w/ truncated image
      [test_freeimage.py] remove extraneous statements
      [test_freeimage] Allow ValueError to be raised
      [skimage.io.tests] add truncated jpg imread tests

Kirill Shklovsky (1):
      Changed the 180 constant in bin calculation to avoid rounding errors

Korijn van Golen (9):
      Partially fulfilled code review.
      Updated docstrings.
      Switched to cnp where applicable.
      Switched to memory view syntax.
      Removed unused import, applied pep8 formatting and corrected a method call
      Removed unused code.
      Renamed to use row/column naming convention.
      Wrapped long lines according to pep8.
      Added unit test to check correct output of HOG on Lena grayscale image, and added feature_vector parameter to disable the .ravel() call on the result.

Kshitij Saraogi (1):
      added support for boolean types in segementation

Kyle Mandli (8):
      Add source related to generation of functionality coverage.
      Modify TASKS to include pointer to coverage docs
      Colored missing text in table
      Add complete matlab image API
      Change look and feel of table, add better handling of table columns
      Change default css to include coverage bar formatting
      Bug fixes and documentation writing
      Fix display of coverage bar

Lee Kamentsky (1):
      Issue #803 order the structuring element by distance to get diagonals processed last

Leena P (3):
      Clarified that Bento Building is an alternative
      Updated novice module tutorial to show images
      Use a simpler operator in the novice tutorial

Leon Tietz (2):
      Updated Image Segmentation tutorial
      corrected plot title

Luis Pedro Coelho (8):
      Fix freeimage_plugin.imsave
      BUG: Correctly handle images with an alpha channel
      ENH Add imread io plugin
      RFCT Use `skimage.utils.convert` for type conversion
      DOC Add myself to CONTRIBUTORS.txt
      TST Remove unneeded test
      ENH Faster perimeter computation
      RFCT Simpler code for initializing array

Malcolm Reynolds (4):
      FIX: number of peaks returned from feature.corner_peaks
      TEST: added test for num_peaks kwarg in peak_local_max.
      FIX: move corner detection outside loop in test_num_peaks.
      FIX: changed xrange to range so Python 3 works.

Marianne Corvellec (24):
      Fixed deprecated threshold.
      Fixed layout (bracket alignment).
      more intuitive description for beginner contributor to get her hands on documentation
      Used  instead of deprecated function.
      Documented another easy way to build scikit-image for virtualenv users.
      Edited after Stefan's comments.
      Okay, I think I got the rst syntax this time...
      Improving docstring + style/convention edits
      Added some warnings for transparency.
      PEP8 compliance!
      One line.
      Clarified that you can pass images of any type into Canny filter.
      Fixed relative threshold: you consider 10% and 20% of the max for range of
      Improved style.
      Always follow the style guide :)
      Included Tony's edits.
      Changed variable names so they actually make sense (followed Tony's
      Added elements to the coffee picture docstring.
      Edited docstring according to comments.
      Tiny formatting edits
      Added deprecation decorator according to discussion.
      Merge branch 'master' of git://github.com/scikit-image/scikit-image into filter_median_dtypes
      Added item to to-do list for version 0.10
      Oops! Typed one instead of the other again! Fixed.

Martin Bergtholdt (1):
      BUG: Windows does not have round in math.h

Matt McCormick (7):
      BUG: Fix testing failures when FreeImage not installed.
      BUG: Fix tests when FreeImage is not installed.
      STY: Align multi-line string statements.
      BUG: Ignore colorconv RuntimeWarning:invalid value encountered in true_divide.
      BUG: Fix OTSU thresholding tests with matplotlib IO plugin.
      TST: Add buffer for threshold tests.
      ENH: Add SimpleITK IO plugin.

Matt Terry (38):
      add color distance functions, ciede2000 tests
      docs
      better cmc implementation
      don't need lab2lch
      add tests. cmc stops naning
      cleaner lch implementations
      fix path
      set testing tolerance
      make naming consistent
      add references
      references
      references
      expose lab/lch conversion
      docs
      pep8 math operators
      use np.rollaxis
      reference formatting
      copy editing docs
      test for rgb2lch roundtrip
      deg2rad > DEG
      skimage attribution convention
      run_module_suite
      docs for deltaE_ciede2000 length scales
      copy editing
      lab2lch/lch2lab for N-D images (w/tests)
      last round of copy-edits
      code cleanup
      tuples are valid inputs
      cleanup
      unneeded asarray
      doc cleanup
      fix table formatting
      spell ciede correctly
      better roundoff handing of dH
      more consistency in naming, better docstring
      r is confusing, use ab instead
      doc quickfix
      comma usage

Matthew Brett (1):
      MAINT: point wheel upload script to terryfy

Matthew Trentacoste (9):
      Luv colorspace conversions
      Adding references regarding Lab and Luv spaces to docstring
      Removing unneeded division checking code
      renaming machineEps to eps to match convention
      flake8-compliant formatting
      Cleaning up docstrings
      docstring changes to correctly format input/output, clarifying XYZ and reference white
      Adding note for Luv colorspace to top docstring
      Merge pull request #1 from stefanv/luv-colorspace

Matěj Týč (25):
      Refactored measure.label, added documentation.
      Added support for 3D arrays
      Refactored Cython stuff with easier code readability in mind
      Cython minor fixes
      Fixed typos + added docstrings and comments
      numpy 1.6 compat fix
      Added some comments + requested reformatting
      Added tests for 3D labelling (that fail)
      Fixed bugs so test pass now
      Mainly typo fixes
      Fixed a typo + cleaned the code semantics
      Redone the background handling = considerable speedup
      Removed declaration of deceased function
      Added and tested proper exception handling concerning array shape
      Introduced the connectivity parameter, regrouped some code
      Fixed typos and made some style fixes
      Made the scan code cleaner and faster (but longer)
      Merge branch 'label3d' of https://github.com/matejak/scikit-image into label3d
      Fixed bugs
      Mended code style
      Fixed typos, added array shape checking
      Streamlined connectivity vs neighbors handling
      Added fix for the strange input array shapes + tests
      Removed negative connectivity support as requested
      Update CONTRIBUTORS.txt

Maximilian Albert (1):
      Fix copy/paste error in docstring.

Michael Hansen (16):
      Added novice submodule after rebase
      Novice module can open images from a url
      Switching to RGBA
      Fixed Python 3 import errors, added alpha and rgba properties
      Trying to get imports working with both Python 2 and 3
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      Fixed import of urlparse and rebase on upstream
      Removed extraneous trys for six imports
      Merged with upstream
      Merged with master
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      Merged with ahojnnes novice
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      Convert incoming novice images to ubyte

Michael Sarahan (3):
      Add subpixel-precision image translation registration function to feature module
      Incorporate fixes for complex images
      improve test coverage.  Use SciPy's fourier_shift for examples.

Michal Romaniuk (5):
      Added SLIC-zero to SLIC and changed SLIC implementation slightly
      Bug fix: removed temporary debug code
      Merged the separate loops for SLIC-zero and SLIC into one, and some minor improvements based on feedback on Github.
      Some changes based on feedback from discussion on Github.
      Added a test for slic_zero and some other changes based on feedback from Github.

Neeraj Gangwar (13):
      added a functionality to fetch images from URL in _novice.py
      added description and exmaple for new functionality
      changed test url in example for the Picture(object)
      changed the method of checking if a string is URL
      added a dummy useragent in Picture class while fetching image
      changed urlparse library from six.moves.urllib_parse
      corrected syntax for error handling while fetching image from
      changed the method of fetching images in Picture class
      removed unnecessary imports from _novice.py
      made minor changes in syntax
      corrected the formula to calculate the location of
      minor changes in syntax in _glcm_loop method.
      added testcases for greycomatrix

Neil (21):
      Remove outdated reference to 'load_plugins'
      Convert from CRLF line endings
      Rework exception handling logic in freeimage_plugin.py
      Fix logic error in _load_library
      replace TemporaryFile with NamedTemporaryFile, for compatability with later python versions
      Only call MultiImage if PIL is available to ensure nose handles this case properly
      Remove temporary files
      Replace old raise form
      Return correct type for as_grey with freeimage
      Pass as_grey to the imread plugins
      Skip pil_plugin tests if PIL isn't available
      Don't rely on pil for plugin_priority test. Skip test if neither pil or freeimage are available
      Revert "Pass as_grey to the imread plugins"
      Remove as_grey from plugins
      Fix bug with 2D images
      Add qt-based imread function
      Only enable imread for new enough sip versions. Warn user if sip is too old
      Try initialise multiple plugins, not just PIL
      Don't return an alpha channel if it's not present. Fix return type of as_grey
      Remove as_grey from qt imread
      Fix axes color order

Neil Yager (25):
      Initial revision of skeletonization
      Added skeleonize example
      Merge remote-tracking branch 'upstream/master' into neil_yager-skeletonize
      Added more objects to skeletonize demo
      Added unit tests to skeletonization
      Added a new unit test
      Updates based on review comments
      Address 2nd round of review comments
      Bug fix: Ignore alpha channel when opening RGBA image as grayscale
      Homography docstring update
      BUG: Fix failing test_homography unit test.
      BUG: Problem importing imread
      NF: Initial commit of greylevel co-occurance (pure python)
      ENH: Compute some standard propertices for GLCMs
      ENH: Added a gallery example of GLCM textures
      DOC: Added latex formulas to docstring
      ENH: Use Cython to compute GLCM
      BUG: Make sure image array is contiguous
      ENH: Updates based on review comments
      BUG: Fix typo and docstrng
      ENH: Address division by zero errors
      BUG: Fixed warning suppression error
      Long lines, remove doctest, division by zero, data type ranges
      Update to docstring
      Fix bug in skeletonize LUT

Nelle Varoquaux (9):
      Renamed scikits.image to skimage
      skimage/version.py should be ignored. Doc has been updated with new module
      Harris corner detector
      Added tests
      Small fixes on the Harris corner detector - documentation, method renaming etc
      Added tests for harris, and small fixes
      Added parameter to Harris corner detector to set the deviation for the gaussian kernel of the harris response computation
      small fixes on the Harris' documentation: added capital letters where needed
      Added Gaël Varoquaux to the corner detector contributors

Nelson Brown (20):
      Add default structuring element to morphology
      Add docstring for default structuring element
      Remove unnecessary whitespace
      Address reviewer comments
      Add gitignore for *.pyd Windows libraries
      Add a newline at EOF
      Add newline at EOF of selem.py
      Add fallback decorator for 3D images
      Add 3d-fallback tests for binary functions
      Add doctrings and comments in response to reviewers
      Add tests for ndimage fallback for 3D white_tophat & black_tophat
      Adjust hanging indentation whitespace in dict comprehension
      Add whitespace following comma
      Change binary morphology functions to output numpy.bool dtype
      Remove repeated comments
      Updated docstrings
      Add test that the 2D results correspond to scipy's
      Fix calls to ndimage to use footprint argument
      Change test of ndimage equivalence to use uint8 test image
      Change ndimage test image to fit uint8 range

Nicolas Pinto (16):
      ENH: move 'util.array_views' to 'util.shape'
      STY: pep8 in util.shape + its tests
      FIX: revert version.py
      STY: fix typos
      STY: move 'util.shape' contributors to CONTRIBUTORS.txt
      ENH: add montage2d with tests
      ENH: add a 'normalize' kwarg to util.montage2d()
      DOC: add example of use of view_as_blocks with kmeans for simple dictionary learning
      FIX: util.montage2d 'normalize' becomes 'rescale_intensity' and defaults to False
      DOC: update view_as_blocks and gabors_from_lena example using comments from @stevanv (PR #138)
      DOC: don't divide by 255 when it's not required ;-)
      DOC: remove dependency on PIL, see PR #138
      DOC: remove <BLANKLINE> nastiness, replace with appropriate doctest directive
      FIX: montage2d should not use rescale_intensity's out_range kwarg
      BUG: fix side-effect in util.montage()
      DOC: "gabors from lena" isn't under construction

Nicolas Poilvert (6):
      initial push of array view codes. tests for block views
      remove empty line in __init__
      more tests on rolling views
      simplifying code for clarity and modifying tests
      make doctests pass with nose
      DOC: add example for util.shape.view_as_{blocks,windows}

Nicolas Rougier (9):
      Remove ticks, add titles and reverse color such that contours are visible
      Switched lena for astronaut
      Removed ticks
      Rescale intensity to enhance image output
      Remove ticks, add titles and reverse color such that contours are visible
      Switched lena for astronaut
      Rescale intensity to enhance image output
      Merge branch 'gallery-cleaning' of github.com:rougier/scikit-image into gallery-cleaning
      Fix style (pep8)

Noah Trebesch (1):
      Fixed correct_mesh_orientation for when spacing!=(1,1,1).

Olivia (12):
      Updated the explanation of the watershed markers example.
      Grammatical error/typo fixed
      Split comment into two rows
      Split comment into two lines
      Added new functionality to remove small holes from images.
      Removed extraneous print statement
      Added new _supported_types function and new line at end
      Changed function name to _check_dtype_supported
      Changed the test file to use expected warnings.
      Moved the comments on labeled arrays to notes section
      Made changes to doctests
      Adjusting spacing of array in doctests

Olivia Wilson (5):
      Altered doctests to correct spacing
      Adjusted white space for doctests
      Added new line
      Merge branch 'remove_small_holes' of https://github.com/oew1v07/scikit-image into remove_small_holes
      Removed trailing whitespace

Olivier Debeir (212):
      fix. gitignore
      add test/
      mv test to tests
      rename using scikits naming conventions
      try to fix rank setup
      rename some test to demo
      add demo all
      add demo all (cont.)
      add demo all (cont.)
      add readme
      add py wrappers
      fix bitdepth in rank.py
      update setup to be compatible with scikits-image
      add percentile mean
      add other rank filters
      add percentile filters - in progress
      add percentile filters
      add bilateral filters pop and meanµ
      add full test
      adapt tests
      add examples - to be cont.
      add readme
      add ref
      add comment to bilateral denoising example
      compare local and global equalise in example
      rename egalise to equalize
      clean-up code
      compare ctmf.median_filter with rank.median
      add comment
      fix percentile autolevel
      fix percentile autolevel
      adapt autolevel example
      fix 16bit percentile
      clean-up code
      fix equalize
      fix other equalize
      add marked watershed example
      add marked watershed example (cont.)
      add marked watershed example (cont.)
      add local test
      remove unused _core
      add exhaustive comparison between 8bit and 16bit filters
      fix 8bit-16bit discepencies
      fix doctest in rank
      remove ascontiguousarray(eimage)
      remoce sr,sc from cores
      remoce selem_num
      add comment
      rename n_se_n to num_se_n etc, removed commented code
      remplace int by Py_ssize_t and fix some doctests
      remplace int by Py_ssize_t and for rank8
      remplace int by Py_ssize_t and for rank16
      remplace emask with is_in_mask function
      add comment
      remplace emask with is_in_mask function in crank16
      group core8,8p and 8b
      delete core8p
      group crank16 and crank16p
      delete core16p
      grou rank16 and rank16b, remove core16b
      find error in autolevel and percentile autolevel (16bit)
      fix error in autolevel and percentile autolevel (16bit)
      add histogram_increment and decrement to core8
      add histogram_increment and decrement to core16
      keep true test in /test, move temporary tests in local
      clean up
      compare ndimage.percentile in demo
      add _apply(func8,func16,...) helper function
      autopep8 sources
      moved example to doc
      removed useless tools.py
      modify CONTRIBUTORS.txt
      __docformat__  removed
      move rank/ into filter/
      cut long lines
      autopep8
      add example modal
      cut long line, autopep8
      remove inlines in pxd
      move setup from /filter/rank to /filter
      remove comments
      remove obsolete comments
      adjust setup
      removed mains
      Merge branch 'master' of https://github.com/scikit-image/scikit-image
      fix. gitignore
      add test/
      mv test to tests
      rename using scikits naming conventions
      try to fix rank setup
      rename some test to demo
      add demo all
      add demo all (cont.)
      add demo all (cont.)
      add readme
      add py wrappers
      fix bitdepth in rank.py
      update setup to be compatible with scikits-image
      add percentile mean
      add other rank filters
      add percentile filters - in progress
      add percentile filters
      add bilateral filters pop and meanµ
      add full test
      adapt tests
      add examples - to be cont.
      add readme
      add ref
      add comment to bilateral denoising example
      compare local and global equalise in example
      rename egalise to equalize
      clean-up code
      compare ctmf.median_filter with rank.median
      add comment
      fix percentile autolevel
      fix percentile autolevel
      adapt autolevel example
      fix 16bit percentile
      clean-up code
      fix equalize
      fix other equalize
      add marked watershed example
      add marked watershed example (cont.)
      add marked watershed example (cont.)
      add local test
      remove unused _core
      add exhaustive comparison between 8bit and 16bit filters
      fix 8bit-16bit discepencies
      fix doctest in rank
      remove ascontiguousarray(eimage)
      remoce sr,sc from cores
      remoce selem_num
      add comment
      rename n_se_n to num_se_n etc, removed commented code
      remplace int by Py_ssize_t and fix some doctests
      remplace int by Py_ssize_t and for rank8
      remplace int by Py_ssize_t and for rank16
      remplace emask with is_in_mask function
      add comment
      remplace emask with is_in_mask function in crank16
      group core8,8p and 8b
      delete core8p
      group crank16 and crank16p
      delete core16p
      grou rank16 and rank16b, remove core16b
      find error in autolevel and percentile autolevel (16bit)
      fix error in autolevel and percentile autolevel (16bit)
      add histogram_increment and decrement to core8
      add histogram_increment and decrement to core16
      keep true test in /test, move temporary tests in local
      clean up
      compare ndimage.percentile in demo
      add _apply(func8,func16,...) helper function
      autopep8 sources
      moved example to doc
      removed useless tools.py
      modify CONTRIBUTORS.txt
      __docformat__  removed
      move rank/ into filter/
      cut long lines
      autopep8
      add example modal
      cut long line, autopep8
      remove inlines in pxd
      move setup from /filter/rank to /filter
      remove comments
      remove obsolete comments
      adjust setup
      removed mains
      small comparison with denoise_bilateral
      resolve conflict
      test noise filter
      fix noise_filter
      fix noise_filter
      doc entropy
      doc noise filter
      modify example entropy
      add local Otsu threshold
      Merge pull request #1 from ahojnnes/rank-filters
      fix: plt.show() removed in doc
      remove local demo from git
      add new tests
      fix: check pop==0 for some kernels
      add: test entropy 8- & 16-bit
      add: test otsu
      add: local Otsu synthetic example
      fix: local otsu doc
      Merge pull request #2 from ahojnnes/rank-filters
      autopep8 for plot_* and rank sources
      fix:wrap lines in tests
      fix:< Py_ssize_t > 0 and & pop
      fix: ref and test_otsu doc
      fix:doctring rank/percentile
      typo
      remove long lines
      fix:remove unused parameters in rank.percentile
      add rank filter sum
      shorten line
      fix:rank.sum docstring
      fix:rank.sum docstring
      fix:rank.sum docstring
      fix:indent
      add:rank.sum_percentile
      add:rank.sum_bilateral
      fix:rank.sum docstrings
      fix:docstring in sum_bilateral
      fix:rtank.__init__
      remove import/blank lines
      fix >80 linelength
      fix entropy example doc, fix figure syntax
      typo
      English corrections in entropy example

Pablo Márquez Neila (4):
      Fix bugs in HOG
      Replaced convolve1d by ad-hoc gradient computation
      Code valid for all image dtypes
      Minor fix

Panfilov Egor (3):
      Merge pull request #1880 from OrkoHunter/fix_doc
      Merge pull request #1885 from OrkoHunter/find_available_plugins
      Merge pull request #1901 from OrkoHunter/test_match_fix_lena

Paul Ivanov (1):
      complete singularization of scikit-image

Pavel Campr (1):
      fix hog.py - orientation and visualization

Phil Elson (3):
      Added gitwash as an option on the makefile.
      Updated the gitwash build.
      Some minor documentation enhancements.

Phil Schaf (1):
      fixed Qt5 compat

Pieter Holtzhausen (44):
      Added CellProfiler sobel and prewitt functions.
      Fixed PEP8 and added numpy documentation.
      Added destination to Makefile and fixed doc system
      Fixes
      Removed followed files
      push script fix
      Push script excludes all git* fix
      Removed generation
      Script to checkout docs repo and commit (ipython based)
      Makefile option for gh-pages generation.
      Update the numpy sphinx extensions from http://projects.scipy.org/numpy/browser/trunk/doc/sphinxext/
      Updated the documentation repo location to the offical destination.
      Tag now a prettier version number from setup.py
      GStreamer and opencv player versions.
      Seeking in video
      Added video duration routines
      First working version of radon and iradon
      Implemented various filters and interpolation methods.
      Test and image dimension checking
      Tutorial
      Created Video base class and handle backend import errors.
      Tutorial edit
      Phantom image example for radon transform
      get_collection method interfacing ImageCollection
      Docstring mods
      Docstring optional
      Tutorial 80 line length
      Fixes
      Probabilistic hough transform
      Testing different accumulator clearing on good lines.
      Merge
      Hough tutorial
      Formatting
      Tutorial work
      Tutorial, adjusted threshold
      Improved line detection
      Added a settle count for similar angles to resolve themselves
      Debugging similar angles
      Merge
      Tutorial save figs
      Fixes
      Removed unused definition
      Fixed opencv reversion to using cv.fromarray, fixed numpy datatype.
      Fixed padding in radon, iradon. Tests for small images.

Pietro Berkes (3):
      Fix LBP variance calculation.
      Fix random seed for test stability.
      No need to take the center of the results twice.

Pradyumna Narayana (6):
      sigma_range is renamed to sigma_color and win_size scales with sigma_spatial
      Added the chnage to comments
      Casted ceil to int such that win_size is an int.
      Supports skimage_range. Displays deprecated warning when used. Two test cases to test this functionality.
      Added deprecated kwargs to TODO.txt
      Corrected a grammatical error in deprecated message. No functional change.

Pratap Vardhan (19):
      CLN: Improve performance of disk and diamond selems
      CLN: Correct dtype typo
      CLN: Remove unused import and add module docstring
      CLN: Minor cleanup - unused imports, pep8
      DOC: Add test functions docstrings
      TST: Test for selem.cube
      ENH: ellipse structuring element
      TST: Test for ellipse structuring element
      DOC: Clean up doc strings
      BUG: Change dtype
      DOC: Change selem.ellipse params
      DOC: Misc spell corrections
      DOC: Example spell corrections
      DOC: Revert to old
      DOC: Update bootstrap sphnix links
      BUG: Buffer dtype mismatch, expected 'Py_ssize_t' but got 'long'
      BUG: Buffer dtype mismatch in _seam_carving.pyx
      MNT: Remove never attained return case under gray2rgb
      TST: Aseert check values for older test_gray2rgb case

Ralf Gommers (47):
      Add Stefan's io.py
      Added a MultiImg class similar to the ImageCollection one, to function as the data structure for multi-image files.
      Add MultiImg to __all__.
      Remove Image class, and implement Stefan's suggestions.
      Cleaned up ImageCollection and added tests for it.
      Now really commit all changes.
      Make as_grey settable, and provide reload function.
      Fixes to MultiImage, added tests for it, plus a test TIF image.
      Rename collection test file and fix it.
      Added dtype keyword to imread plus a test for that.
      Add dtype parameter to MultiImagei and ImageCollection.
      Move docs to correct place inside class.
      Change "Additional Parameters" to "Other Parameters".
      Fix link in top corner to always point to main page.
      Add a few docstrings.
      Add support for MPL plots in the docs.
      Update tasks, link to docstandard.
      Remove "Functions" links/headings.
      Clean up title of API reference.
      Use same bullet level for functions and classes.
      Fix up API generation.
      Add build dir to .gitignore.
      Change import to form used everywhere else.
      Add documentation for rgb2hsv.
      Implement hsv2rgb.
      Add note on loss of precision.
      Extract the type and shape checks into a common function.
      Add RGB to XYZ bidi conversion.
      Add docstrings for rgb2xyz and xyz2rgb.
      Add description of the API to the module docstring.
      Convert matrix lists to arrays, add RGB CIE color space.
      Bug fix for imread so images that are already grey-scale do not get flattened.
      Add NTSC color space.
      Clean up the colorconv module docstring.
      Remove NTSC color space.
      Add a function to convert from any color space directly to any other space.
      Remove explicit checks of image type and NaN, and correct dtype descriptor.
      Add color spaces work to contributors file.
      Add note on assumed data range.
      Make "make test" work.
      Add make coverage command.
      Add some docs for 'make coverage'.
      Minor changes to Tony Yu's palette_is_grayscale.
      Fix the description of make coverage.
      Fix link to issue tracker.
      Remove note about patent issues with marching squares, was resolved on list.
      Correct calling of coverage plugin, no spurious output anymore.

Raphael Ackermann (1):
      changing docstring to make clear that only 2-D images are currently supported

Rebecca (8):
      making in-place addition modification suggested in issue #1089
      replacing lena in examples and most tests
      adding name to contributors
      renamed gabors
      removed lena reference
      modified array
      Modified docstring
      removed typo that caused test error

Riaan van den Dool (2):
      GDAL plugin added.
      Forgotten contribution by riaanvddool: skimage.io plugin - GDAL

Rishabh Raj (31):
      Adds functionality to generate ipython notebooks for the gallery examples
      Fixes for making code more pythonic, follow conventions
      added Notebook class for creating IPython notebooks, better name for main object variable
      fixes use of camel case
      added newline at end of file
      moves everything related to generating the IPython notebook, to a different file
      updated import statement for creating ipython notebooks, better name in place for jsonify
      adds documentation
      removes redundant passing of an extra parameter
      Fix PEP8
      Fix camel case, more descriptive function name
      Fix typo, better function name and docstring summary
      Fix PEP8, line lengths for comments
      Add __all__ to specify modules which should be imported
      Add default value for adding cell, better naming for type of value
      Remove sample.ipynb, template now inline
      Add and improve docstrings, JSON dumping and duplicate removal generalised
      Improve docstring for add_cell
      Fix to incorporate a more generalised duplicate removal and JSON dump to string
      Improve docstring for consecutive duplicate removal, more readable
      Remove extra spaces
      Remove fetch_key, an overkill
      Add string options to the docstring
      More pythonic
      Add class to __all__, remove duplicates now private
      Remove segment_number, directly change the last cell
      Fix PEP8, more readable
      More generic Notebook class
      More pythonic if statements
      Add warning, consistent use of quotes, correction in docs
      Remove use of redundant variables, cleanup

Robin Wilson (8):
      Fixed typo in docs README
      Added quantile_threshold option to canny edge detection
      Raise ValueError if quantile thresholds outside of [0, 1] - and updated docs
      Added test for invalid quantile inputs
      Renamed keyword arg to use_quantiles
      Fixed indentation
      Fixed tests to use assertRaises as a function not a context manager
      Added make.bat file for building docs on Windows

SemiColon Warrier (1):
      updated local_binary_pattern documentation

Sergey Karayev (4):
      Fixed RGB<->XYZ conversion.
      flake8 on colorconv.py and removed unneeded .copy()
      Merge pull request #1 from amueller/xyz_fix
      added rgb2lab test

Sergio Pascual (2):
      Add deprecation warning to 'data.lena()'
      Suggest 'skimage.data.astronaut' as alternative test image

Shaun Singh (3):
      Update docstring for profile_line
      Clean up profile-line docstring
      Fix spacing for inline comments

Siva Prasad Varma (15):
      used plt.subplots everywhere
      changed colorbars
      changed to fig.colorbar
      changed to plt.subplots()
      fixed plot in plot_glcm.py
      made plot loop more readable
      changed to plt.subplots
      changed to fig.subplots_adjust()
      changed to fig.subplots_adjust()
      changed to fig.subplots_adjust()
      changed to fig.subplots_adjust()
      changed to fig.subplots_adjust()
      changed f to fig
      changed f to fig in applications
      changed io to ax

Stefan van der Walt (1404):
      Initial framework for scikits.image.
      Add Sphinx documentation.
      Credit contributors.
      Pull in LPI filters and add image reading (via PIL for now).
      Refactor LPI filtering.
      Fix and test filter API.
      Merge branch 'master' of git://github.com/npinto/scikits.image
      Move data files to common directory.
      Move data directory.
      Move data directory.
      Merge branch 'master' of git at github.com:stefanv/scikits.image
      Fix paths in colorconv.
      Add Hough transform.
      Add Makefile.
      Correctly label targets as phony.
      Update contributors.
      Add link to mailing list.
      Local licenses override.
      Add ignores.
      Update ignores.
      Fix houghtf docs.
      Add numpy-sphinx docs extensions.
      Add API generation and fix documentation build.
      Script to automatically build and upload docs.
      Add license info to numpy sphinx extensions.
      Generate cleaner API reference with autosummary tables.
      Update license in docs.
      Docs: clean up front page.
      Docs: include license directly.
      Docs: add header to table of contents.
      Docs: update install and overview.
      Docs: add inheritance diagram extension.
      Fix docstring formatting.
      Minor fixes to patch.
      Remove executable flag.
      Add SURF and SIFT feature file reader.
      Add projective transformation / homography.
      Fix markup in docs.
      Add 3rd clause to license (thanks to David Warde-Farley for noticing the omission).
      Only update gh-pages branch when committing new docs.
      Add documentation on how to contribute to the SciKit.
      Merge branch 'homography'
      Update meta-data for PyPi.
      Instructions on how to make a release.
      Remember to generate docs when releasing. Also update version in setup.py.
      Grab version number from setup.py for docs.
      Move test data to global data dir and fix test.
      Update version to 0.2dev in setup.py.
      Rename collections module.
      Update tasks list.
      Add collections to io.__init__.
      Change mail-archive link to Nabble.
      Rename collection test file.
      Add only_directives.py and plot_directive.py from numpy.
      Remove dependency on only_directives.
      Set plot directory prefix.
      Plot_directive: copy source files to output.
      Shorten and comment show_collection example.
      Clean up output of plotting directive:
      When pushing docs to github, rename _images to images.
      Remove completed task: plots in docs.
      Update license phrase for plot_directive.
      Merge remote branch 'colbert/master'
      Update .gitignore for whole scikit.
      Update setup to build scikits.image.opencv.
      Only build OpenCV extensions when source has been modified.
      setup: Remove dependencies -- easy_install doesn't always know when
      Clean up and fix setup.
      Update build instructions.
      Fix references scikits.gpu -> scikits.image.
      Add setup.py in scikits/image.
      Fix typo in opencv setup.
      Clean up long lines and whitespace in opencv_backend.
      OpenCV: clean up long lines and whitespace.
      Allow unit tests to be run when OpenCV is not available.
      Update contributors.
      Expand OpenCV installation instructions.
      Fix importing of opencv when libraries are not installed / extensions are not compiled.
      Ignore build directory.
      Merge branch 'master' of github.com:stefanv/scikits.image
      Update contributors.
      Merge branch 'docs' of http://github.com/rgommers/scikits.image
      Clean up opencv namespace and imports.
      Remove self-referencing import.
      Remove task completed by Ralf.
      Clean up scikits.image namespace.
      Docs: In API generator, also recognise extensions.
      Update tasks list.
      Add MANIFEST.in for building source package.
      Include auto-generated API files in distribution.
      Fix _libimport when libcv is not available.
      Unify libcv loader under different platforms.
      Update tasks list and release instructions.
      Merge with Chris's branch.
      Fix bug in libimport.
      Add package __version__.
      Merge branch 'color' of http://github.com/rgommers/scikits.image
      Add default version file.
      Merge branch 'master' of http://github.com/sccolbert/scikits.image
      Add dist to gitignore.
      Generate new API only when the current version of the library is installed, otherwise revert to pre-generated files. [Patch by Helge Reikeras]
      Fix opencv for systems without libraries installed.
      Merge branch 'coverage' of http://github.com/rgommers/scikits.image
      Plugin framework for image I/O.
      Merge Chris's cv docs updates.
      Merge branch 'docs' of http://github.com/sccolbert/scikits.image
      cv: Fix function signatures.
      cv: Add references section for adding URL to docs.
      cv: Update Cython generated C-files.
      Add new docs API files.
      Add tests for IO plugins.
      Restore plugin registry after testing.
      Document _call_plugin.
      Remove plugin functionality from io.py.  Implement setting default plugins and querying the plugins available.
      Merge branch 'master' of http://github.com/tonysyu/scikits.image
      Add Tony as a contributor.
      Check that scikit is built before generating docs for upload.
      docs: Backup generated files before switching to docs branch.
      Merge plugin branch.
      Add QT4 image display plugin.
      io: qt4 plugin: allow multiple images to be displayed. Keep track of Windows so that they don't get deallocated before they are closed.
      Add Chris's plugin utility converter and wxpython plugin.
      Rather use global than list to track app.
      io: qt4: Use prepare_for display. Add demo snippet.
      Fix q4thread typo.
      Reorganise plugin infrastructure.
      Second part of I/O infrastructure reorganisation.
      Import PIL plugin using load_plugin.
      Add shortest path algorithm.  Refactor Cython build.
      Merge branch 'io' of git://github.com/sccolbert/scikits.image
      Minor text cleanups.
      Add tests for prepare_for_display.
      Add tests for window manager.
      Fix subtle rstrip bug.
      Fix subtle rstrip bug.
      Fix subtle rstrip bug.
      Describe plugins using .ini files, so that we don't need to import them to know their capabilities.
      Merge branch 'coverage' of git://github.com/rgommers/scikits.image
      io: Clean up 'available' according to Chris's review.
      Merge branch 'io'
      io: Allow plugin names that don't end on _plugin.
      Merge branch 'spath'
      doc: Add analysis API.
      build: Fix c extension.
      spath: Fix indexing errors.
      io: Add plugin descriptions.
      spath: Allow importing even when not compiled.
      io: Track meta-data on plugins.
      io: Add table of available plugins to module docstring.
      analysis: Add as subpackage.
      docs: Update API.
      io: Combine use_plugin and load_plugin.
      doc: Update API.
      Update tasks.
      io: Add plugin .ini files to setup as data files.
      io: Add setup files for colour mixer.
      io: Document colour mixer.
      io: Add tests for colormixer.
      Rename analysis -> graph.
      Remove Cython-generated C-files.
      doc: Update github push script.
      Align image to top so that pixel tracking is accurate.
      Add image stack.
      Merge branch 'io' of git://github.com/sccolbert/scikits.image into io
      Add basic tests for _histograms.
      io: Fix Qt import accidentally removed.
      io: Combine qt imports.
      Update tasks list.
      Add scikits.image.test.
      Add image viewer script.
      Remove `imsave` from matplotlib until fixed in their source.
      Add user guide and documentation on plugins.
      io: Add _app_show for matplotlib plugin.
      Merge branch 'doc' of git://github.com/tonysyu/scikits.image
      doc: Update dependencies.
      Add scripts for launching without setuptools.
      Merge branch 'io' of git://github.com/sccolbert/scikits.image into io
      io: Move plugin tests to io/tests.
      Merge branch 'threadingio' of git://github.com/sccolbert/scikits.image into io
      Merge branch 'threadingio' of git://github.com/sccolbert/scikits.image into io
      io: Fix one more instance of cdivision.
      io: Throw exception when qt plugin can't be loaded.
      io: Add imsave using PIL.
      Merge branch 'io'
      docs: Update API.
      Update tasks.
      Merge remote branches 'colbert/master' and 'colbert/luttables' into chris_merge
      doc: Update API.
      Set version to 0.2.
      Update development status to 'beta'.
      setup: Add MANIFEST.in.
      Revert accidental change out of version.py
      Document problem with 'python setup.py sdist upload'.
      setup: Only check for new Cython file if Cython executed successfully.
      setup: Fix Cython build.
      Fix MANIFEST.
      setup: Fix author field not to be tuple.
      Release 0.2.1 to fix build issues.
      Update release info now that setup is fixed.
      Update tasks.
      Fix with statement for 2.5.
      io: Add null-plugin to simplify testing.
      io: Fix doctests.
      io: Remove unneccesary import.
      Merge branch 'compatibility2.5' of git://github.com/tonysyu/scikits.image
      Update TASKS.
      Fix MANIFEST.in after report by Lev Givon.
      Set version to 0.2.2 to release a bugfix sdist.
      Clean up setup script.
      Setting version to 0.3dev.
      Update TASKS.
      Merge branch 'damian-morphology' of git://github.com/deads/scikits.image into damian
      morphology: Wrap long lines.
      Add Zachy Pincus's trace_path routine for finding a low-cos path between a given start and end point.
      trace_path: PEP7 fixes.
      trace_path: Add tests.
      trace_path: reformat docs
      Add Zachary Pincus to contributors list.
      spath: Remove debug prints.
      Restore version file.
      spath: Add Cython definition file to allow external use.
      docs: Multiple columns in API reference.
      docs: Update API.
      Cython division fixes. Ignore version file after build.
      docs: Add project title to header.
      docs: Fix typo.
      Add easy way to grab a logger.
      docs: Update api.
      Add docs for logger.
      graph: heap: Cast known integer after division.
      graph: spath: Fix documentation indentation.
      morphology: Clean setup.
      graph: mcp: Clean up long lines. Fix tests.
      graph: Fix typo in heap docs. Wrap long line.
      Merge speed fixes to heap.pyx from Zach.
      Update links to CellProfiler code.
      Add Zachary's freeimage wrapper.  Fix Ctypes loader to use NumPy.
      Add description file for freeimage plugin.
      Add imread and imsave wrappers to freeimage plugin.
      Wrap long lines in freeimage plugin.
      In freeimage plugin, fix order of parameters to 'write'.  Provide colour images to 'write' as CxMxN arrays.
      Add ctypes API for freeimage plugin for use on 64-bit platforms.
      In freeimage plugin, use array_interface for array construction.
      Try to load both libfreeimage and libFreeImage.
      Fix freeimage loading via array interface.
      Use c_void_p to handle (*BYTE) in freeimage plugin.
      In freeimage plugin, set DummyArray to writable, use correct dtype.
      Found shorthand for constructing dtype str (thanks, Zach!).
      Update links to OpenCV docs.
      Clarify dependencies.
      Add gitwash documentation.
      Add imshow to PIL plugin.
      Allow PIL plugin to read grey-level images with alpha layer.
      More accurate docstring for imread.
      Merge branch 'master' of git://github.com/sccolbert/scikits.image into chris
      DOC: Add CellProfiler relicensing announcement.
      ENH: Refactor ImageCollection to allow more general file loading schemes.
      DOC: Add place to document API changes.
      ENH: Add imread_collection to plugin framework.
      Ignore vim swap files.
      Fix link to api changes in docs.
      Merge branch 'master' of git://github.com/luispedro/scikits.image
      Modify numpy cimport order to satisfy Cython 0.13.
      BUG: Hough failed on provision of angles argument.
      Update Git docs with link to Fernando's post.
      More updates to dev guidelines based on Fernando's post.
      API: Remove dtype parameter from imread and collections. Remove as_grey from collections.
      ENH: Add rgb2grey.
      ENH: Add a free standing test for freeimage.
      Merge branch 'io_cleanup'
      Merge branch 'imagecollection'
      Merge branch 'james_turner-fits_plugin'
      ENH: Update Cython imports.
      BUG: PIL tests should set plugin at setup.
      DOC: Add explanation to test plugin.
      BUG: Build fixes for Windows 7, Python 2.6 [patch by Amit Aides].
      DOC: Note that grey-level morphological operations only support flat structuring elements [provided by Ingrid Fabris-Rotelli].
      BUG: Fix Windows 7 build fix.
      DOC: Update dependency to Cython 0.13.
      ENH: Find libcv.so.2.1 in addition to just libcv.so.
      DOC: Wrappers are now for OpenCV 2.1.
      ENH: Fix cvUndistort2 according to OpenCV 2.1 calling convention.
      ENH: Add test for cvUndistort2 with new intrinsics provided.
      ENH: Add parameter checking for cvCalibrateCamera2.
      ENH: Add opencv wrapper for FindFundamentalMat.
      BUG: Include default parameters for cvFindFundamentalMat.
      BUG: Fix status shape in cvFindFundamentalMat.
      BUG: OpenCV checks shape of memory provided.
      BUG: Compare ``method``, not ``status``.
      BUG: Status should be 2-dimensional.
      ENH: Add test framework for cvFindFundamentalMat.
      DOC: Note about Ubuntu + cv + Atlas problem.
      BUG: Remove status parameter that causes OpenCV 2.1 to complain.
      BUG: Fill status with zeros. Squeeze dimensions of F.
      ENH: Complete test for cvFindFundamentalMat.
      DOC: Better document test for cvFindFundamentalMat.
      Merge branch 'OpenCV_2.1'
      Merge branch 'cvFindFundamentalMat'
      Update tasks.
      Fix Cython build on Windows [patch by Christoph Gohlke].
      Merge branch 'mcp_infinite' of ssh://github.com/zachrahan/scikits.image
      ENH: Allow stand-alone execution of ctmf test suite.
      ENH: Remove unnecessary C defines.
      ENH: Stylistic changes.
      DOC: Update median_filter doc to numpy format.
      ENH: Clean up whitespace and remove class in ctmf tests.
      ENH: Execute median filter without mask array by default.
      DOC: Give credit to Lee Kamentsky and Thouis Jones.
      Merge constant time median filter by Lee Kamentsky and Thouis Jones.
      BUG: Fix building under Python 3.
      BUG: Fix mcp tests.
      BUG: Cython build: Open file correctly under Python 3 [patch by Nadav Horesh].
      BUG: Python 3 compatibility fixes by Christoph Gohlke.
      BUG: More Python3 fixes by Christoph Gohlke.
      BUG: Do not break on import if PIL isn't available.
      BUG: Fix filename handling under Python 3.
      BUG: Do not rely on PIL by default.
      ENH: Also scan /home/stefan/lib for freeimage.so.
      DOC: Update tasks.
      BUG: Fix sed in docs build script.
      DOC: Update tasks.
      ENH: morphology: Add connected components labelling.
      Merge remote branch 'dfarmer/dfarmer-filters-canny'
      DOC: Add Dan Farmer as contributor.
      ENH: Add faster numpy hough version as well as a Cython version. Add appropriate setup.py files to build the Cython module. Fix hough tests.
      Update Chris's contribution.
      BUG: Import PIL.Image correctly. Skip tests if not available.
      DOC: morphology: Document connectivity in ccomp.
      Merge branch 'ccomp'
      ENH: Minor clean-ups to edge-detection patch.
      DOC: Add Pieter Holtzhausen as a contributor.
      Merge branches 'holtzhau_edge_filters' and 'ccomp'.
      BUG: Correctly convolve integer and floating point arrays.
      ENH: graph: Fix typo.
      Merge remote branch 'zach/FreeImage64'
      Merge remote-tracking branch 'emmanuelle/tv_denoise'
      DOC: Update contributors.
      Merge Dan Farmer's Windows platform updates to the NumPy ctypes library loader.
      DOC: Update contributors.
      Merge Tony Yu's tv denoise fixes.
      Allow test_sift to be ran independently.
      _hough_transform: Type fix.
      Update contributors.
      Merge documentation build system fixes by Pieter Holtzhausen and Tony Yu.
      Update contributors.
      DOC: Update coverage instructions.
      ENH: Add default target for docs Makefile.
      ENH: Ignore .coverage output.
      DOC: Add new project URL to setup.py.
      DOC: Add gitwash generator code.
      DOC: Use the correct suffix when building gitwash docs.
      DOC: Add more prominent link to gitwash docs.
      DOC: Rewrite project overview.
      DOC: Update gitwash generated docs.
      TST: Test both Cython and Python versions of the hough tf.
      DOC: Add missing space.
      Remove outdated push_github script.
      DOC: Save plots with tight bounding boxes.
      DOC: Setup plot_directive.
      DOC: Update hough docstring.
      DOC: Also copy plot files to html doc output.
      ENH: Add hough transform plot.
      ENH: Update gitignores.
      Merge Neil's fixes, including those for py3k.
      Update contributors.
      DOC: Retheme docs, credit IPython authors.
      Merge Tony's documentation enhancements.
      DOC: Shrink table for better layout.
      DOC: Update contributors.
      DOC: Turn gh-pages into phony target.
      DOC: Fix repo address in installation instructions.
      DOC: Minor cleanups.  Strike through completed items on TODO list.
      DOC: Deploy current docs to scikits-image.org/docs/dev.
      Merge coverage states into main .css.
      DOC: Update contributors.
      DOC: Add coverage table to Makefile.
      DOC: Complete coverage bar.
      DOC: Ignore generated coverage table.
      Use absolute path for scikits.image.data_dir.
      Correctly detect missing FreeImage.
      Update license to credit team.
      Merge pull request #17 from emmanuelle/change_repo_name
      DOC: Describe usage of different types (first draft).
      DOC: Fix typo in data types description.
      Use only snake image.
      Adjust orange snake brightness. Add --no-plot flag.
      Add SVG logo with text and Makefile to convert to png.
      Match scikits.image directory structure.
      Merge Tony's logo generator.
      Match logo filename used in website.
      Also produce a scaled down version of the logo.
      Add logo to generated documentation.
      Merge branch 'logo'
      DOC: Note that data type document is still preliminary.
      Merge branch 'dtype_info'
      DOC: Update Tony's contributions.
      DOC: Update release instructions.
      Add summed area table, using Nicolas Pinto's suggestion for a pure
      DOC: Update sat docs after feedback from Brian Holt.
      Merge branch 'summed_area_table'
      Merge radon transform by Pieter Holtzhausen.
      Merge pull request #22 from mandli/master
      ENH: Rename sat and sat_sum to be in-line with CV community convention.
      ENH: Expose data module. Currently just provides scipy's lena in a more convenient location.
      DOC: Update tv example to use data module. Rewrap some text.
      Merge Emmanuelle's work on documentation examples / build.
      Merge pull request #24 from tonysyu/doc_tweaks
      Add Cython homography implementation.  Not well optimised yet.
      ENH: Minor clean ups of data loader.
      DOC: Update contributors.
      Merge branch 'data_loader'
      ENH: fast_homography: 20x speed improvement.
      BUG: fast_homography: Fix mirror mode.
      Fix mirror mode. Update docstrings. Update tests
      BUG: fast_homography: Fix constant mode.
      Import fast homography as transform.fast_homography.
      Merge radon/fast_homography.
      ENH: Use fast homography instead of PIL's rotate. Minor cleanup.
      Add drawing module with Bresenham line.
      ENH: Use new draw module to construct HOG image.
      ENH: Minor setup.py cleanups.
      BUG: Remove numpy 2 usage of .astype.
      Merge pull request #33 from amueller/median_filter_minor
      ENH: Add dtype converters.
      DOC: Fix dtype converter docstring typos.
      BUG: Fix typo in dtype converter integer list.
      DOC: Warn about OpenCV deprecation.
      TST: Make LPIFilter tests operate on a smaller image, since they take long to execute, and cameraman.png is now bigger than before.
      DOC: Move radon example to gallery.
      DOC: Use correct location of blank_image in example generator.
      DOC: Update Makefile's clean target.
      BUG: Always return contiguous array from colour conversion.
      BUG: fast_homography should ensure input data is contiguous.
      BUG: Fix fast_homography test based on new lena.
      BUG: Use local lena data file in tv_denoise test.
      DOC: Update contributor howto.
      ENH: Add img_as_ubyte.
      Merge branch 'dtype_converters'
      BUG: Add 'util' subpackage.
      Merge pull request #35 from amueller/color_converters_dtype
      Merge pull request #29 from amueller/canny_usability
      Merge pull request #36 from amueller/dtype_minor
      PKG: Allow make clean to run twice in a row.
      PKG: Remove both .so and .pyc files in find . -name "*.so" -o -name "*.pyc" | xargs rm -f.
      BUG: Make test file non-executable.
      BUG: Add 'color' subpackage.
      BUG: Adopt lena example to use new data module.
      Merge pull request #38 from amueller/dtype_minor
      BUG: Fix default log-level.
      Merge Andy's plugin cleanups require to build the Debian package.
      PKG: Remove empty placeholder files.
      PKG: Do not build Cython files on clean.
      Merge pull request #40 from emmanuelle/watershed
      Add drawing module with Bresenham line.
      ENH: Use new draw module to construct HOG image.
      BUG: Add 'draw' and 'feature' sub-modules.
      Merge pull request #28 from stefanv/hog
      Merge pull request #27 from stefanv/radon
      Merge furhter improvements to histogram of gradients.
      Merge pull request #12 from holtzhau/hough
      Merge pull request #10 from holtzhau/video-io
      Qt-based image reading.
      DOC: Update contributors.
      Merge video I/O.
      BUG: Fix watershed tests.
      BUG: Use integer image values as expected in tv denoise test.
      DOC: Move Hough tutorial to examples gallery.
      DOC: Remove PIL dependency.
      DOC: Release notes for 0.3.
      DOC: Update release notes.
      PKG: Update version number for release.
      DOC: Update doc url in release notes.
      PKG: After release, update version to 0.4dev.
      DOC: Update release instructions (run setup.py register to login to pypi).
      Merge pull request #42 from riaanvddool/gdal_plugin
      DOC: Typo in release notes.
      DOC: Fix broken link in watershed docs.
      DOC: Do not rebuild the doc versions index automatically.
      PKG: Also include .pxi files in the source package.
      PKG: Add python 3 and operating system tags to setup.py.
      BUG: Untabify video.py.
      Merge pull request #48 from cgohlke/patch-5
      Merge pull request #47 from cgohlke/patch-4
      Merge pull request #49 from cgohlke/patch-6
      Merge pull request #50 from cgohlke/patch-7
      Merge pull request #52 from cgohlke/patch-15
      Merge pull request #51 from cgohlke/patch-8
      Merge pull request #53 from cgohlke/patch-1
      Replace .npy files by .npz.  Smaller, and avoids pickling problems on Python 3.
      Use .npz test data in opencv tests.
      Merge npz fixes from Christoph Gohlke.
      Fix another instance of an .npy file.
      Merge npy -> npz transition.
      Remove OpenCV wrappers for 0.4 release.
      BUG: Fix scikits.image.test.
      DOC: Update contributors.
      Remove empty gitwash directory in order to fix submodule.
      DOC: Add gitwash as a submodule.
      DOC: Use correct repo URL in gitwash.
      DOC: Rewash the git docs.
      Merge pull request #56 from cgohlke/patch-1
      Merge pull request #59 from cgohlke/patch-2
      Merge pull request #58 from cgohlke/patch-3
      Merge pull request #63 from amueller/matplolib_plugin
      Merge remote-tracking branch 'nelle/skimage' into nelle_varoquaux_skimage_rename
      PKG: Move remaining files that changed since skimage rename.
      DOC: Add Nelle to contributors.
      Rename package to ``skimage``.
      DOC: Update repo address.
      PKG: Import skimage instead of scikits.image in changes made since rename.
      DOC: Update gitwash.
      DOC: Minor spacing fixes in docstrings.
      Merge Emmanuelle's port of CellProfiler's medial axis skeletonization.
      Merge pull request #67 from NeilYager/neil_yager-alpha
      ENH: Add binary convex hull computation.
      ENH: Add coins image.
      DOC: Better docstring for coins.
      ENH: Pre-allocate coordinate corner array.
      ENH: Improve convex_hull execution speed.
      ENH: Only perform convex hull on select candidate pixels. Gets us a factor 30 speedup.
      Merge pull request #77 from NeilYager/neil_yager-homography-docstring
      PKG: Move io to _io (closes #39).
      Merge pull request #73 from tonysyu/fix-ctmf
      DOC: Update tasks list.
      DOC: Add url for convex hull algorithm overview.
      ENH: In test, do not compute hull twice.
      BUG: Allow test suite to pass with scipy < 0.9.
      Update test data set descriptions.  Convert 'coins' to grey-level image.
      DOC: Update tasks with a link to a free qhull implementation.
      Merge pull request #76 from stefanv/coins
      ENH: Rename convex_hull to convex_hull_image to allow later clean API addition of convex_hull with labels.
      Merge pull request #75 from stefanv/convex_hull
      ENH: Use hashes of .pyx files to see whether re-compilation is needed or not.
      Merge pull request #74 from stefanv/cython_hash
      Merge pull request #78 from emmanuelle/doc_canny
      DOC: See also to skimage.sobel instead of ndimage.sobel.
      ENH: Add verbose test function.
      BUG: data: Fix chessboard file name.
      Merge pull request #79 from zachrahan/mcp-64bit
      DOCS: Tighten borders around image in GLCM example.
      Merge branch 'neil_yager-greycomatrix'
      DOC: Use convex_hull_image instead of convex_hull.
      DOC: Improve example gallery layout.
      BUG: Import find_contours from correct location.
      ENH: Whitespace cleanup.
      REF: Convert find_contours to Cython.
      BUG: Import imread from renamed _io module.
      DOC: Update contributors.
      DOC: Add contour finding example.
      STY: Clean up docstrings.
      DOC: Bold headers for Notes, Reference, etc. sections in reference docs.
      STY: Remove unused imports.
      DOC: Fix team name in license and docs.
      Merge pull request #82 from zachrahan/find-contours
      DOC: Fix some scikits.image mentions.
      DOC: Script to generate contributors.
      DOC: 0.4 release notes.
      BUG: Python 3 fixes.
      Release 0.4.
      Move to 0.5 development.
      DOC: Update release process.
      Merge pull request #84 from cgohlke/patch-1
      Merge branch 'master' of github.com:scikits-image/scikits-image
      BUG: Wrong distribution name.
      DOC: Update installation instructions.
      Merge pull request #89 from cgohlke/patch-4
      BUG: Fix selective plugin loading. Provide function to inspect current plugin configuration.
      ENH: Use dtype scaling routines in qt and pil plugins.
      ENH: Tie imshow and _app_show together.
      Merge pull request #90 from stefanv/plugin_load_fix
      DOC: Update Cython dependency.
      Merge pull request #96 from emmanuelle/glcm
      Merge pull request #97 from cgohlke/patch-1
      DOC: Update 0.4 release notes (wrong gallery url).
      Ignore auto examples.
      BUG: Correctly load bi-level images via PIL.
      Merge pull request #102 from tonysyu/pil-big-endian
      ENH: Allow choice between 4 and 8 neighbor mode.  Improve documentation.
      Merge pull request #104 from tonysyu/pil-big-endian
      DOC: Add index of tutorials to user guide.
      DOC: Add Emmanuelle's SciPy India segmentation tutorial.
      BUG: Dispatch threads to process entire image.
      ENH: Add background labelling.
      DOC: Update contributors.
      DOC: Explain why bincount is used to histogram integer arrays.
      DOC: Rename equalize example to match function name.
      DOC: Move around plot statements so that all axes are labeled.
      Merge branch 'tony-exposure'
      DOC: Display thresholding example with gray colormap.
      DOC: Add script to plot GitHub pull-request activity.
      DOC: Better x-axis labelling of PR plot.
      BUG: Fix OpenCV array conversion in video loader.
      DOC: Add script to randomly link to a gallery image.
      DOC: Generate correct link to gallery and build javascript before gh-pages upload.
      DOC: Ignore random.js generated file.
      BUG: Edge detection routines returned images with invalid ranges [0-255] but dtype float.
      Merge pull request #115 from stefanv/edges_float_conversion
      ENH: Use nearest neighbor interpolation by default in matplotlib plugin.
      Merge pull request #112 from zachrahan/freeimage-64-bit-cleanup
      STY: Whitespace cleanup freeimage patch.
      Merge pull request #118 from 'zachrahan-freeimage-new-metadata'
      ENH: When displaying images using matplotlib, use the grey colormap by default.
      Merge pull request #110 from tonysyu/peak-detection
      Merge pull request #99 from cgohlke/patch-2
      ENH: Add contrast stretching to histogram equalization example.
      DOC: Fix some typos in plot_equalize example.
      ENH: Cleanups as suggested by Tony.
      PKG: Rename scivi to skivi.
      PKG: Remove version file that should be auto-generated.
      Merge pull request #121 from tonysyu/dtype-convert-updates
      PKG: Add back version file, but with version unbuilt-dev.
      BUG: Fix import of harris.
      Merge pull request #123 from tonysyu/doc-user-guide-dtypes
      Merge pull request #122 from tonysyu/doc-css-changes
      Merge pull request #107 from tonysyu/rescale-intensity
      DOC: Show version of skimage that was used to build example.
      Merge pull request #126 from michaelaye/master
      DOC: Ignore all auto generated results.
      Merge pull request #127 from tonysyu/module-docstrings
      DOC: Expand on Tony's module summary.
      Merge remote-tracking branch 'tonysyu/versions-in-sidebar' into tonysyu-versions-in-sidebar
      DOC: Refactor Javascript for listing doc versions to only have the version in one place.
      PKG: Update release instructions to update doc versions javascript.
      DOC: Use absolute path to versions javascript.
      Merge PR #128 from 'tonysyu/versions-in-sidebar'.
      DOC: Add arrow to indicate current doc version.
      Safely check doc version in javascript.
      DOC: Link logo to main webpage. Add navigation side bar.
      DOC: Fix Other Parameters section.
      ENH: Fix dtype conversion before hsv2rgb and rgb2hsv.
      ENH: Add grey2rgb.
      Merge pull request #129 from tonysyu/minor-build-fixes
      Merge pull request #131 from tonysyu/versions-sidebar-tweaks
      DOC: Fix outdated docstring.
      Merge pull request #132 from tonysyu/doc-dtype-tweaks
      DOC: Update dependencies.
      DOC: Update installation instructions to match those on web page.
      DOC: Add FreeImage as optional dependency.
      Use matplotlib as the default display/read plugin.
      ENH: Add show() to PIL plugin to support more consistent code over different plugins.
      DOC: Import hog from skimage instead of scikits-image.  Rescale the output histogram to be more visible.
      Merge PR #135 from 'bdholt/enh-hog' after rebasing.
      DOC: Fix formatting typos in HoG example.
      BUG: Fix FreeImage tests.
      DOC: Add FreeImage as optional dependency.
      DOC: Import hog from skimage instead of scikits-image.  Rescale the output histogram to be more visible.
      DOC: Fix formatting typos in HoG example.
      STY: Whitespace cleanup.
      BUG: Fix FreeImage tests.
      Merge PR #134 from 'zachrahan/freeimage-linuxload'
      Merge pull request #139 from cgohlke/patch-3
      Merge pull request #141 from emmanuelle/user_guide
      Merge pull request #142 from tonysyu/fix-readme-file
      Merge pull request #138 from poilvert/array_views
      Merge pull request #143 from tonysyu/array-view-docstring
      BUG: Fix data range in exposure tests.
      BUG: Use PIL as default imread plugin, since MPL imread is still somewhat broken.
      Merge pull request #148 from amueller/doc_error_message
      Merge PR #147 from 'tonysyu/fix-gallery-sizes
      DOC: Update tasks.
      PKG: Move README.md to README.rst.
      Merge pull request #151 from tonysyu/cleanup-tasks
      Merge pull request #152 from cgohlke/patch-1
      Merge pull request #156 from tonysyu/fix-tifffile-test
      Merge pull request #155 from cgohlke/patch-4
      DOC: Add image colorization to tasks.
      DOC: Add v0.5 release notes.
      PKG: Update version to 0.5.
      PKG: Update doc versions.
      PKG: Move development to 0.6dev.
      DOC: Update release instructions.
      DOC: Fix typo in 0.5 release notes.
      DOC: Update release instructions.
      PKG: Change gitwash submodule git remote so that ShiningPanda can pull that repo.
      TST: Try to get Jenkins build working via Tox.
      ENH: Remove annoying version.py.
      Merge pull request #157 from tonysyu/fix-plugin-call
      PKG: Fix version import to allow build on 3.2.
      BUG: Fix imports for Python 3.2.
      Merge pull request #160 from stefanv/python3.2-fixes
      BUG: Fix typo in montage--xrange is no longer supported.
      BUG: Correctly fix zip import for Python 3.2.  2to3 simply removes the itertools import line.
      ENH: Add freeimage as fallback io plugin, to allow buildbot to read test images.
      ENH: Allow forced copy of data upon type conversion.
      Explicitly use float64 instead of float.
      ENH: Do not use global root logger state in skimage.
      Merge pull request #171 from emmanuelle/modif_radon
      Merge pull request #176 from tonysyu/peak-fixes
      DOC: Add Yaroslav Halchenko to contributors.
      Merge pull request #177 from ahojnnes/draw-polygon-alt
      BUG: Provide bresenham as backward compatible name for line.
      Merge pull request #179 from neurodebian/master
      Merge pull request #169 from stefanv/local_logging
      Merge pull request #181 from tonysyu/fix-logging-py26
      Merge pull request #178 from ahojnnes/adaptive-thresholding
      BUG: Remove import of non-existent module.
      ENH: Implement fast coordinate transformations.
      BUG: Fix incorrect import.
      DOC: Add CSS for LaTeX math.
      Merge pull request #187 from tonysyu/morph-cleanup
      Merge pull request #186 from holtzhau/radon_fixes
      Merge pull request #182 from tonysyu/threshold-example
      BUG: Fix Python 3 syntax error.
      STY: Cleanups after Tony's review.
      STY: Move swirl to _warp_zoo so we can add more transforms later.
      BUG: Ensure that the appropriate I/O plugin is used in the test suite.
      BUG: Fix plugin import on systems without PIL or FreeImage.
      TST: Increase radon transform threshold to make test more robust.
      STY: Minor cleanups.
      Merge pull request #164 from cournape/bento_build
      BUG: Temporary workaround for shape offset parameter in radon transform.
      ENH: Better formulation of _stackcopy.
      Merge pull request #100 from tonysyu/skimage-template
      BUG: Fix internal import in template matching.
      BUG: Correctly import _feature in Python 3.
      Merge pull request #189 from tonysyu/fix-stack-copy
      Merge pull request #191 from cgohlke/patch-1
      BUG: Fix FreeImage Py3k compatibility [patch by Almar Klein].
      BUG: Catch all exceptions, including NotImplementedError, when checking for number of CPUs in multi-processing.
      Merge pull request #192 from tangofoxtrotmike/master
      Merge pull request #183 from ahojnnes/measure
      BUG: Remove debugging print statement that broke Py3k.
      Merge pull request #198 from zachrahan/mcp-index-fix
      Merge pull request #193 from cgohlke/patch-1
      PKG: Update release instructions--add version to bento.info.
      BUG: Fix PIL test on big endian.
      Merge pull request #200 from cgohlke/patch-2
      BUG: Fix background labelling for case when (0, 0) belongs to the background.
      Merge pull request #202 from cgohlke/patch-3
      DOC: Provide correct link for tifffile plugin.
      Merge pull request #196 from tonysyu/fix-plugin-init
      BUG: Use PIL as the default image loader.
      ENH: Allow resetting the plugin state.
      ENH: Add Structural SIMilarity (SSIM) image comparison.
      ENH: Add SSIM gradient.
      ENH: Rename ssid to structural_similarity to avoid confusion with self-similarity features.
      DOC: Example of structural similarity.  The example is currently broken because structural_similarity does not auto-detect dynamic range.
      DOC: In ssim example, use correct dynamic range in algorithm and when displaying results.
      ENH: Automatically determine dynamic range in ssim if not specified.
      PKG: Rename _ssim to _structural_similarity.
      STY: Wrap long line.
      ENH: Promote as_windows to a utility function.
      PKG: Rename as_windows to view_as_windows.
      BUG: Remove double import of find contours.
      BUG Remove merge artefact.
      BUG Fix invalid import of structural_similarity.
      BUG: Fix structural similarity to use new signature for view_as_windows.  Remove bad gradient check.
      Merge pull request #190 from tonysyu/plot2rst
      DOC: Add 0.6 release notes.
      PKG: Update version to 0.6.
      PKG: Update doc versions.
      PKG: Start 0.7 development cycle.
      PKG: Use different delimiter in contributors list.
      DOC: Update release instructions.
      Merge pull request #205 from amueller/pep8
      BUG: Allow rgb2grey to be called on grey-level images.
      TST: Check number of input angles in radon tf.
      Merge pull request #214 from cgohlke/patch-1
      Merge pull request #216 from ahojnnes/perimeter
      PKG: Update Debian packaging instructions.
      DOC: Add SciPy as a runtime dependency.
      DOC: Update tasks.
      Merge pull request #220 from thewtex/freeimage_error
      Merge pull request #222 from thewtex/freeimage_error_2
      Refactor geometric transforms.
      Merge pull request #225 from lct123/ImageSegTut
      Merge pull request #223 from jni/jni-concatenate-image-collection
      Merge pull request #228 from thewtex/color_numpy_warnings
      Merge pull request #219 from wilsaj/remove-nose-message
      Merge pull request #197 from ahojnnes/projection
      DOC: Fix typos in see-also references.
      Merge pull request #249 from ahojnnes/peak-local-max
      Merge pull request #252 from gatagat/patch-regionprops
      Merge pull request #240 from tonysyu/rename-duplicates
      BUG: Fix broken import in ctmf.
      DOC: Update development guidelines.
      DOC: More coding conventions.
      Merge pull request #244 from pcampr/master
      Merge pull request #239 from ahojnnes/texture
      Merge pull request #206 from amueller/felsenzwalb
      Merge pull request #256 from NeilYager/skeleton_fix
      BUG: Remove uses of xrange for py3 compatibility.
      DOC: Update development guidelines from PR 255 feedback.
      Merge pull request #250 from ahojnnes/approx-polygon
      Merge pull request #258 from tonysyu/imshow-filename
      BUG: Remove print statement that caused py3 tests to fail.
      DOC: Add intersphinx so we can refer to numpy, scipy, sklearn docs.
      Merge pull request #229 from tonysyu/qtmpl-viewer
      BUG: Workaround to unit tests failing if PyQt4 not available.
      Merge pull request #261 from emmanuelle/fix_import_glcm
      PKG: Update release notes, plot_pr.
      Merge pull request #260 from ahojnnes/convert-bool
      Merge pull request #266 from tonysyu/viewer-fix-hard-exit
      Merge pull request #267 from tonysyu/viewer-adapt-dtype
      Merge pull request #215 from tonysyu/morph-reconstruction
      Merge pull request #257 from ahojnnes/shared
      Merge pull request #269 from tonysyu/viewer-fix-doc-and-test
      Merge pull request #270 from tonysyu/cython-0.15-compat
      Merge pull request #274 from ahojnnes/doc-update
      Merge pull request #273 from ahojnnes/draw-file-deletion
      Merge pull request #276 from tonysyu/reconstruction-fix
      Merge pull request #272 from ahojnnes/template-contiguous
      Merge pull request #259 from emmanuelle/fix_random_walker
      Merge pull request #277 from emmanuelle/fix_tests_draw
      Merge pull request #275 from ahojnnes/warps
      Merge pull request #288 from tonysyu/bento-update
      Merge pull request #287 from ahojnnes/morph
      Merge pull request #290 from tonysyu/qtmpl-lineprofile
      Merge pull request #289 from tonysyu/qtmpl-collection-viewer
      Merge pull request #291 from tonysyu/sphinx-import-warnings
      Merge pull request #295 from tonysyu/doc-cleanups
      BUG: Add missing imports in transform module.
      Merge pull request #297 from ahojnnes/piecewise-affine
      Merge pull request #300 from tonysyu/viewer-qapp-fix
      TST: Add Travis configuration based on nipy's.
      Merge pull request #303 from jaberg/local_logger
      Merge pull request #308 from ahojnnes/doc-font-size
      Merge pull request #305 from ahojnnes/setup-import-warning
      Merge pull request #313 from ahojnnes/travis-tests
      Merge pull request #301 from tonysyu/viewer-user-guide
      Merge pull request #310 from ahojnnes/morph
      TST: Invoke Travis nosetests differently.
      Merge pull request #315 from tonysyu/doctest-plot-fix
      ENH: Bool dtype conversion speed improvements.
      Conversion to bool: threshold on half dtype max.
      Merge pull request #322 from JDWarner/random_walker_docfix
      Merge pull request #323 from emmanuelle/bug_rw
      BUG: Fix homography.
      PKG: Remember to check that random.js file is correctly generated.
      Merge pull request #326 from amueller/cleanup_imports
      Merge pull request #332 from cgohlke/patch-8
      Merge pull request #327 from cgohlke/patch-1
      Merge pull request #328 from cgohlke/patch-3
      Merge pull request #329 from cgohlke/patch-4
      Merge pull request #330 from cgohlke/patch-5
      Merge pull request #331 from cgohlke/patch-7
      Merge pull request #334 from tonysyu/fix-apigen
      DOC: Fix typo in release instructions.
      Merge pull request #335 from ahojnnes/doc-theme
      Merge pull request #337 from tonysyu/example-gallery-fix
      DOC: Credit Johannes for website layout.
      Merge pull request #343 from ahojnnes/warps-openmp-fix
      Merge pull request #345 from ahojnnes/docs-favicon
      Merge pull request #342 from ahojnnes/domain-name
      Merge pull request #347 from ahojnnes/doc-version-link
      DOC: Rename skimage.org to scikit-image.org.
      DOC: Remove confusing paragraph about dtype range.
      Merge pull request #359 from ivanov/singular-scikit
      DOC: Clean up one last skimage -> scikit-image.
      PKG: Print something when bento check passes.
      DOC: Fix pip command.
      Merge pull request #293 from ahojnnes/resample
      Merge pull request #362 from ahojnnes/perimeter-test
      Merge pull request #365 from ahojnnes/ref-fix
      Merge pull request #364 from ahojnnes/scharr
      BUG: Allow RGB images to be passed to grey2rgb.
      Add is_rgb and is_gray.
      Merge pull request #349 from ahojnnes/bilateral-filter
      PKG: Fix bento build.
      ENH: Add motion blurred clock image.
      Update README to Markdown.
      Merge pull request #373 from luispedro/imread_io_plugin
      BUG: WindowsError not defined under Linux.
      Merge pull request #376 from JDWarner/unify_peak_finder_backend
      Merge pull request #388 from ahojnnes/umfpack-warning
      DOC: Minor clean-ups of equalization demo.
      Minor code cleanup.
      Merge pull request #393 from ahojnnes/namespace-fix
      Merge pull request #368 from tonysyu/imread-url
      Merge pull request #402 from ahojnnes/perimeter-fix
      Merge pull request #395 from jni/jni-slic-bw
      Merge pull request #321 from ahojnnes/interest-points
      Merge pull request #242 from blink1073/adapthist
      Merge pull request #398 from ahojnnes/polynomial-order-default
      Merge pull request #405 from blink1073/feature-travis-py3k
      Merge pull request #372 from ahojnnes/tv-filter
      Merge pull request #406 from blink1073/bug-adapthist-py3k
      Merge pull request #412 from ahojnnes/draw-cython-directives
      PKG: MSVC build fixes by Christoph Gohlke.
      Merge pull request #420 from ahojnnes/random-seed
      Merge pull request #425 from ahojnnes/shared
      Merge pull request #433 from odebeir/fix_rank_doc
      Merge pull request #424 from ahojnnes/ssize_t
      Merge pull request #421 from thewtex/SimpleITK_io
      Merge pull request #435 from ahojnnes/cnp
      Merge pull request #434 from ahojnnes/ssize_t-note
      PKG: MSVC build fixes by Christoph Gohlke.
      Merge remote-tracking branch 'cgohlke/patch-2' into msvc_build_fix
      Merge pull request #438 from ahojnnes/warp-fix
      Merge pull request #431 from sciunto/houghcircle
      DOC: Use brighter color in circular Hough demo.
      BUG: Recast label output to satisfy ndimage on 32-bit platforms.
      Merge pull request #444 from jmargeta/MSVC_uint16_fix
      STY: Minor cleanup.
      Merge pull request #450 from ahojnnes/hough-circle
      DOC: Fix typo in remove_small_objects docstring.
      Merge pull request #454 from ahojnnes/release-announcements
      Merge pull request #455 from thouis/ctmf_unaligned_segv
      STY: PEP8 for logo.
      Merge pull request #457 from tonysyu/logo_refactor
      Merge pull request #390 from tonysyu/canvastools
      Merge pull request #460 from jni/raise-type
      Merge pull request #462 from ahojnnes/deprecated-doc-string
      Merge pull request #463 from ahojnnes/pdf-docs
      Merge pull request #441 from spotter/color_deconv
      Merge pull request #473 from ahojnnes/feature-refs
      Merge pull request #501 from ahojnnes/skeletonize
      Merge pull request #500 from JDWarner/fix_preserve_rw_input_shape
      Merge pull request #503 from ahojnnes/exposure
      Merge pull request #494 from ahojnnes/release-instr
      Merge pull request #496 from ahojnnes/label-performance
      Merge pull request #488 from sciunto/minor_docfix
      Fix Emmanuelle Gouillart's name in the contributors list.
      Refresh contribution guidelines.
      Add upstream / username remotes.  Illustrate rebase.
      ENH: Add `chelsea` to dataset.
      BUG: Fix integer type in Felzenwalb cython.
      Merge pull request #515 from tonysyu/fix/imread-url-bug
      Merge pull request #527 from JDWarner/tempfix_use_old_cython
      Merge pull request #530 from npinto/patch-2
      Merge pull request #540 from ahojnnes/namespace
      Merge pull request #440 from ahojnnes/fitting
      Merge pull request #550 from ahojnnes/matching-spelling
      Merge pull request #566 from ahojnnes/bscript
      Merge pull request #565 from ahojnnes/doc-fix
      Merge pull request #564 from ahojnnes/warp-example
      Merge pull request #568 from ahojnnes/circle
      Merge pull request #567 from josteinbf/radon-reconstruction-circle
      Merge pull request #538 from ahojnnes/examples-travis
      Remove reference to Bento hookfile.
      Merge pull request #581 from ankit-maverick/subtract
      Merge pull request #485 from tonysyu/image_label2rgb
      Merge pull request #577 from JDWarner/add_pad
      Merge pull request #601 from ahojnnes/corner-eps
      Merge pull request #612 from applio/master
      Remove 2to3.
      Add six compatibility module.
      Remove is_str and replace by isinstance(..., six.string_types).
      Correctly check for string type.
      Fix zip.
      Fix function code.
      Fix iterator.
      Fix label casting for numpy 1.7.
      Fix im_class access in Python 3.
      Fix im_self access in Python 3.
      Fix imports under Python 3.
      Use int32 for labelling.
      Add  to public API.
      Space code more readably in transform.
      Ensure PIL tests run with PIL.
      Do not run flakes on test files -- we have too many custom formatted array.
      Merge pull request #596 from josteinbf/radon-projection-center
      Merge pull request #642 from sciunto/convex_hull_ex
      Fix Bento build.
      Merge pull request #614 from tonysyu/doc/local-binary-pattern
      Merge pull request #646 from ahojnnes/travis-bento
      Fix casting in ransac example.
      Merge pull request #645 from josteinbf/iradon-sart-docstring
      Merge pull request #611 from mkcor/install_pip-requirements
      Add javascript for locating latest stable version.
      Highlight current doc version.
      Shorten Bento installation instructions and fix broken links.
      Environment variable is called WAFDIR.
      Properly fix WAF environment directory. Tell users to download source (not binary).
      Merge pull request #626 from ahojnnes/rank-fixes
      Merge pull request #627 from ahojnnes/travis-examples
      Merge pull request #666 from jni/slic-scaling
      Merge pull request #654 from josteinbf/radon-square-center
      Merge pull request #674 from ahojnnes/region-props
      Merge pull request #677 from ahojnnes/radon-example
      Merge pull request #676 from ahojnnes/regionprops-fixes
      Merge pull request #688 from ahojnnes/regionprops-bugfix
      Merge pull request #694 from ahojnnes/moments
      Merge pull request #696 from ahojnnes/foerstner-ref-fix
      Merge pull request #695 from ahojnnes/memoryviews
      Merge pull request #698 from mrterry/deltae_docs
      Remove logger in favor of warnings.
      Remove unnecessary import of Warning.
      Merge pull request #697 from ahojnnes/memoryviews
      Remove leftover debug statement.
      Remove Image wrapper from io.  Improve tags handling.
      Add tests for image tags.
      Merge pull request #638 from sciunto/contribute
      Merge pull request #711 from sciunto/minordoc
      Merge pull request #721 from ahojnnes/travis
      Add skip parameter to view_as_windows.
      Merge pull request #750 from ahojnnes/misc
      Update PR plotter.
      Correctly plot cumulative sum.
      Test whether labels are left alone in label2rgb.
      Merge pull request #773 from ahojnnes/scipy-0.13rc
      Merge pull request #759 from jni/supress-umf
      Add labels to releases.
      Merge pull request #7 from tonysyu/pr_plots
      Merge pull request #675 from sciunto/ellipseht_heapq_fix
      Merge pull request #756 from ahojnnes/morph
      Simplify output handling for binary morphology.
      Merge pull request #749 from ahojnnes/gsoc-experimental
      Fix overflow in NumPy 1.7 as suggested by Julian Taylor.
      Restore @ahojnnes's overflow test.  Correctly assign out argument.
      Test that output argument is utilized.
      Test that output argument is correct.
      Revert to >0 for determining binary values. Document that input should be binary.
      Pre-allocate memory for convolution and re-use for output, if no user-override specified.
      Allocate less memory, if possible.
      Update docs to match output.
      Tweak markup of docstring.
      Provide stable doc version by naming docs directory 0.9.x, e.g.
      Always use uint8 for binary view.
      Merge pull request #783 from ahojnnes/draw
      Merge pull request #784 from ahojnnes/deprecations
      Update TODO.txt after deprecations.
      Note deprecations under API changes.
      Rewrite contributor lister in Python.
      Add contributors to 0.9.
      Update author order for 0.9 release notes.
      Dictionary interface to preserve backward compatibility for RegionProps.
      Fix Python 3 list handling.
      Fix Py3 exception in GTK plugin.
      Merge pull request #792 from stefanv/gtk_py3
      Update contributors script to count by date.
      Contrib script now shows PRs and merges.
      Update 0.9 release notes with new features.
      Update versions for 0.9.0 release.
      Update version in docs.
      Correctly determine version number from module.
      Fix markup error in marching cubes docs.
      Update gh-pages instructions in RELEASE.txt.
      Mark BRIEF and Censure as experimental in release notes.
      Update docversions correctly for 0.9.x.
      Update manifest not to include gh-pages in docs.
      Set version numbers for 0.10 development cycle.
      Update website update instructions for release.
      Correctly determine dev version for doc building.
      Get rid of that inherited 's' once and for all.
      Merge pull request #795 from amueller/fix_felsenszwalb_minsize
      Merge pull request #796 from ahojnnes/warp-fix
      Make six an external dependency.
      Add python-six as Travis dependency.
      Install six from PyPi.  Travis version is too old.
      Format color conversion docstrings.
      Merge pull request #809 from ahojnnes/rank-setup
      Update pip requirements.
      Merge pull request #811 from ahojnnes/doctests
      Add coverall support
      Execute coveralls without flags.
      Change to test directory before running coveralls
      Ignore setup. No need for tmp dir any longer.
      Fix invocation of nosetests.
      Merge pull request #822 from JDWarner/fix_freeimage_metadata_test
      Merge pull request #798 from matttrent/luv-colorspace
      Merge pull request #823 from JDWarner/novice_py26_compatible
      Ignore main sections of library code
      Merge pull request #644 from josteinbf/phase-unwrap
      Merge pull request #825 from ahojnnes/coverage
      Merge pull request #848 from ahojnnes/warps
      Merge pull request #800 from orieux/wiener-deconvolution
      Merge pull request #854 from almarklein/update-mcp
      Merge pull request #859 from radioxoma/threshold
      Merge pull request #824 from tonysyu/fix/hough-circle-regression
      Merge pull request #840 from ahojnnes/estimator-params
      Merge pull request #834 from ahojnnes/orb
      Replace Cython's round with a VC2010 compatible version
      Merge pull request #865 from jni/profile-line
      Use frombytes instead of fromstring on PIL images, since the latter is now deprecated
      Add '--all' flag to placify newer versions of git
      Fix Image.frombytes for older version of Python/PIL
      Only run coverage for Python 3.
      Exit with status zero under Python 2
      Travis-ci doesn't like 'if-else'?
      Merge pull request #888 from ankit-maverick/issue885
      Merge pull request #894 from blink1073/viewer_docstring
      Merge pull request #903 from vighneshbirodkar/dog
      Merge pull request #906 from ahojnnes/peak-fix
      Merge pull request #890 from ahojnnes/matching
      Make doc/gh-pages executable
      Merge pull request #922 from juliantaylor/doc-improve
      Merge pull request #915 from vighneshbirodkar/blob_det
      Merge pull request #923 from ahojnnes/doctests
      Merge pull request #916 from neerajgangwar/master
      Merge pull request #924 from ahojnnes/restoration
      Merge pull request #937 from sciunto/doc
      Merge pull request #938 from sciunto/draw
      Merge pull request #939 from sciunto/py3
      Merge pull request #935 from neerajgangwar/fix-greycomatrix
      Remove debug print in contribs
      Use nearest neighbor interpolation and a gray-level color map for plotting feature matches
      Merge pull request #958 from ahojnnes/numpydoc
      Merge pull request #961 from ahojnnes/video
      Merge pull request #967 from pberkes/fix/lbp-variance
      Merge pull request #955 from neerajgangwar/add-test-greycomatrix
      Fix parameter documentation in corner_peaks (closes #901)
      Do not allow view_as_* to return copies (closes #866)
      Take a softer approach and raise a warning when a non-contiguous input is provided
      Update view_as_windows docstring
      Print warnings while testing
      Fix typo in docstring
      Merge pull request #975 from jni/isrgb-depr
      Merge pull request #974 from jni/slic-depr
      Distinguish between executable Python and Python version for apt-get
      Merge pull request #977 from ahojnnes/todo-0.10
      Fix coverage config specification in nosetests invocation
      Use coveralls instead of python-coveralls
      Rename _remove_consecutive_duplicates to _squash_repeats
      Update Notebook docstring
      Add unit tests
      Add tests for notebook. Remove python_to_notebook utility function.
      Update plot2rst to generate notebooks
      Merge pull request #936 from sharky93/ipython-notebookcreation
      Make another pass through contributing guidelines
      Merge pull request #982 from sciunto/contributing
      Merge pull request #830 from astrofrog/use-astropy
      Add all_warnings context manager
      Update shape tests to use all_warnings
      In all_warnings context manager, return warnings for examination
      Move all_warnings to skimage._shared.utils
      Copy list of modules since it may change during iteration
      Add some PEP8
      Fix broken link to origin apigen source
      Merge pull request #986 from vighneshbirodkar/blob_example
      Replace broken _import_tools from numpy with Git version
      Print versions of software used in Travis output
      Remove deprecated --use-mirrors from pip calls
      Rename heading -> header
      Don't fail if .matplotlib dir already exists
      Merge pull request #998 from ahojnnes/python34
      Always interpret provided shapes as int. Add a note on using output_shape for color images.
      Add unit test to catch non-integer output shape bug
      Clarify regionprops output
      Move label to measure module (with deprecation). Warn that background parameter will default to 0 in v0.12.
      Merge pull request #8 from JDWarner/safe_int_casting
      Move implementation of label to measure submodule
      Fix imports and suppress deprecation warnings
      Use safe_as_int to check provided output shape
      Update bento build
      Move safe_as_int into utils
      Fix broken and circular imports
      Mention which parameters to label are optional. Simplify label deprecation.
      Merge pull request #1008 from ahojnnes/coverage
      Merge pull request #1005 from JDWarner/int_idx_patch
      Merge pull request #1014 from tonysyu/update-EPD-link
      Merge pull request #1015 from maxalbert/master
      Avoid circular imports
      Correctly apply catch_warnings()
      Fix markup in DEPENDS
      Update MANIFEST to include all test data
      Use more recent NoNorm API
      Correctly compare mask to None
      Ignore auto-generated notebooks in Git
      Fix median filter example
      Use same order as matplotlib for PySize/PyQT
      Merge pull request #1033 from jni/label_image
      Only import numpy after dependency check
      Merge pull request #1039 from kdmurray91/fix_freeimage_segfault
      Add OSX wheel generation instructions and upload script
      Update wheel URL
      Get tags from current repo
      Indicate waiting time before wheels become available
      Remember to update stable docs link.  Check timestamp on wheels before posting.
      Merge pull request #1059 from nlsn/exclude-win-libs
      Merge pull request #1032 from tonysyu/better-rescale-intensity
      Merge pull request #1095 from ahojnnes/denoise-tests
      Merge pull request #1098 from ahojnnes/tox
      freeimage: Correctly handle saving uint16 images (closes gh-1101)
      Merge pull request #1116 from rebeccaroisin/inplace_addition
      Merge pull request #1121 from Cadair/doc_theme_tweaks
      Merge pull request #1124 from blink1073/master
      Remove leftovers of old logger
      Merge pull request #1115 from JDWarner/enh_marching_cubes
      Merge pull request #1113 from jucoste/move_canny
      Merge pull request #1132 from ahojnnes/win-hist
      Fix indexing bug in freeimage plugin
      Fix Big Endian check
      Merge pull request #1114 from pelson/gitwash
      Merge pull request #1043 from sciunto/joblib
      Merge pull request #1144 from blink1073/master
      Merge pull request #1167 from ahojnnes/inv-comb
      Merge pull request #1168 from ahojnnes/rot-center
      Merge pull request #1162 from blink1073/improve-pil-error-messages
      Fix incorrectly documented non-optional argument
      Expose point_in_poly and move to measure module
      Merge pull request #1081 from SiggyF/fedor-setup-installrequires
      Merge pull request #1174 from jni/deconv-shape-fix
      Allow modification of LineTool handle style
      Add more complete documentation for widget callback
      Invoke linetool with properties to increase coverage
      Reminder to remove previous installations when upgrading on Windows
      Improve my grammar
      Also add handle style to ThickLineTool
      Fix class names
      Merge pull request #1130 from capitanbatata/more-illuminants
      Merge pull request #1176 from ahojnnes/ransac
      Merge pull request #1193 from pratapvardhan/morph-selem
      Merge pull request #1175 from blink1073/update-setup
      Merge pull request #1189 from blink1073/fix-pil-plugin
      Merge pull request #1198 from blink1073/travis-sh
      Merge pull request #1202 from jni/badge
      Merge pull request #1199 from blink1073/use-pypi-tifffile
      Merge pull request #1204 from blink1073/travis-streamline
      Merge pull request #1206 from tonysyu/fix/canny-deprecation
      Update error message for no plugins
      Print correct version when dependency is not satisfied
      Merge pull request #1171 from jjhelmus/unwrap_phase_3d_bug
      Merge pull request #1200 from blink1073/multiimage_tifffile
      Use equal_nulp for more robust test comparison
      Update edge detectors to produce values in [0, 1] and clean up tests
      Document threshold scaling
      Remove prints in test suite
      Merge pull request #1222 from sciunto/add_shape_draw2
      Rename and deprecate filter module to prevent shadowing of built-in keyword
      Check for deprecation on import is problematic.  Rather just check that filter can be imported normally.
      Fixes according to PR feedback
      Do not render filters documentation twice
      Ensure import of old filter namespace raises warning
      Filter won't be imported twice--check for its deprecation correctly
      For `filter` submodule, explicitly set warning registry location
      Add debug print to failing assert
      Make sure warning is raised upon import
      Simplify warning test
      Merge pull request #1226 from blink1073/show-travis-build-versions
      Remove build status until Travis-CI yields stable builds
      Merge pull request #705 from emmanuelle/userguide
      Merge pull request #1237 from emmanuelle/filters_rank
      Merge pull request #1249 from arve0/fix-doc-build-virtualenv
      Merge pull request #1251 from arve0/fix-github-logo-icon
      Merge pull request #1247 from emmanuelle/gallery_edges
      Merge pull request #1254 from matejak/label3d
      Merge pull request #1252 from sciunto/readme
      Merge pull request #1250 from emmanuelle/seealso
      Merge pull request #1253 from blink1073/travis-venv
      Webhook to report Travis-CI status on Gitter
      Merge pull request #1268 from pratapvardhan/morph-selem
      Merge pull request #1267 from blink1073/fix-travis-retry
      Merge pull request #1263 from blink1073/fix-load-imgnum
      Merge pull request #1275 from blink1073/use-wheelhouse-simpleitk
      Merge pull request #1285 from blink1073/fix-travis-setup
      Merge pull request #1281 from blink1073/fix-imshow
      Merge pull request #1273 from blink1073/increase-pip-timeout
      Merge pull request #1286 from blink1073/python3-doc-build
      Merge pull request #1288 from pratapvardhan/morph-selem
      Merge pull request #1292 from emmanuelle/numpydoc
      Merge pull request #1293 from blink1073/streamline-travis
      Merge pull request #1294 from blink1073/streamline-travis
      Improved example notebook generation
      Fix warnings generated by gallery examples
      Fix integer division in CLAHE
      Fix docstring markup errors
      Raise an error that was suppressed to please Travis-CI
      Merge pull request #1277 from emmanuelle/api_layout
      Merge pull request #1299 from ahojnnes/block
      Merge pull request #1051 from ahojnnes/bicubic-interp
      Add Gitter badge
      Merge pull request #1259 from emmanuelle/ug_transform
      Merge pull request #1303 from ahojnnes/unwrap
      apigen: Fix function detection in Python 3
      Merge pull request #1307 from ahojnnes/unwrap-rand
      Merge pull request #1337 from jni/fix-label2rgb
      Merge pull request #1338 from blink1073/imshow-fixes
      Merge pull request #1339 from jni/slic-doc
      Merge pull request #1318 from blink1073/pyqt5
      Merge pull request #1310 from jni/morpho-update
      Merge pull request #1348 from jjhelmus/fix_unwrap
      Merge pull request #1319 from ahojnnes/geom-est
      Merge pull request #1280 from jni/array_order
      Merge pull request #1352 from jni/data-types
      Merge pull request #1356 from blink1073/travis-build-docs
      Merge pull request #1371 from blink1073/fix-reference-links
      Merge pull request #1079 from blink1073/fix-colorconv
      Merge pull request #1377 from blink1073/test-against-pil
      Fix gh-pages under Python 3. Make robust against running twice in a row.
      Update README and installation instructions
      Mention license type
      Merge pull request #1320 from blink1073/pr/1057
      Merge pull request #1077 from buzmakov/master
      Merge pull request #1066 from msarahan/PhaseCorrelation
      Add @msarahan to CONTRIBUTORS.txt
      Add missing 0.10 release notes
      Add 0.11 release notes
      Merge pull request #1398 from stefanv/0.11_release_notes
      Update version to 0.11.0
      Fix DEPENDS markup
      Fix another formatting error in DEPENDS
      Set master branch to 0.12
      Update tagging instructions for release
      Merge pull request #1411 from ahojnnes/rans-fix
      Allow clear_border to operate on labeled images
      Merge pull request #1452 from blink1073/fix-pkg-resources
      Remove coverage table.  Add document with release notes and download instructions.
      Merge pull request #1453 from blink1073/only-index-tiff-gif
      Merge pull request #1408 from stefanv/tagging
      Merge pull request #1461 from matthew-brett/fix-wheel-upload
      Add spacing in novice model to make docs render correctly
      Pick up Cython functions when building docs API
      gh-pages: Read skimage version from __init__.py
      Merge pull request #1475 from gustavla/slic-fix
      Correctly refer to combined installation and release notes
      Sync doc theme with website
      Fix reference to release notes
      Merge pull request #1494 from arve0/reset_button_crop_plugin
      Improve test robustness
      Add in-place operation
      Merge pull request #1504 from ahojnnes/denoise_bilateral_fix
      Merge pull request #1485 from emmanuelle/model_binary
      Merge pull request #1488 from danielwe/blob-float-sigma
      Merge pull request #1511 from grlee77/ssim_ndim
      Merge pull request #1517 from ahojnnes/parallel-decoration
      Merge pull request #1519 from ahojnnes/nogil
      Merge pull request #1493 from cowlicks/chunk-map
      Merge pull request #1534 from michaelaye/patch-1
      Merge pull request #1542 from michaelaye/patch-2
      Remove logo (moved to branding repo)
      Simplify installation instructions
      Consistently use relative imports
      Update contribution guidelines to recommend relative imports and releasing the GIL
      Subtle wording improvement to test setup docstring
      Remove invalid link to table of contents (closes #1593)
      Fix and clean up anti-aliased line drawing (closes #1563)
      Rename starting point of line
      Merge pull request #1600 from michaelpacer/fix_docs_sprint
      Merge pull request #1596 from david1koeller/novice_picture_fixes
      Merge pull request #1597 from library111/DocumentationChanges
      Merge pull request #1584 from JDWarner/enh_make_clean
      Merge pull request #1606 from blink1073/fix-setup-numpy
      Merge pull request #1610 from blink1073/pyqt-test
      Merge pull request #1612 from blink1073/fix-manage-plugins
      Merge pull request #1613 from blink1073/better-test-warnings
      Merge pull request #1614 from blink1073/handle-pip-upgrade
      Merge pull request #1622 from jjhelmus/check_build_sprint
      Merge pull request #1626 from JDWarner/fix_rank_test
      Merge pull request #1628 from JDWarner/sphinx_linkcode
      Merge pull request #1633 from JDWarner/fix_travis_networkx
      Merge pull request #1634 from JDWarner/imsave_no_squeeze
      Add alpha layer support to gray2rgb
      Add utility to rename functions
      Rename nl_means_denoising to denoise_nl_means
      Merge pull request #1656 from cgilon/contribute-docs-fix-link
      Merge pull request #1695 from MartinSavc/adding_makefile_params
      Merge pull request #1701 from blink1073/fix-draw-polygon-1700
      Merge pull request #1697 from blink1073/deploy-docs
      Merge pull request #1702 from blink1073/fix-doc-build
      Merge pull request #1709 from blink1073/windows-docs
      Merge pull request #1711 from blink1073/docs-push-again
      Merge pull request #1713 from blink1073/docs-fix-take-3
      Merge pull request #1715 from blink1073/docs-build-token
      Merge pull request #1716 from blink1073/docs-build-again
      Merge pull request #1719 from blink1073/another-docs-build
      Merge pull request #1720 from blink1073/another-token-attempt
      Merge pull request #1722 from blink1073/add-doc-deploy-notes
      Merge pull request #1741 from escorciav/peak-local-max-output-type
      Merge pull request #1753 from blink1073/fix-numpy1.10
      Merge pull request #1750 from jni/slic-param-scan
      Ensure gallery pages are always generated
      Merge pull request #1746 from sciunto/piwik
      Only iterate over available region properties
      Remove dict comprehension for Python 2.6 compatibility
      Merge pull request #1805 from jni/fix_slic
      Merge pull request #1803 from JDWarner/mesh_orientation_cleanup
      Merge pull request #1811 from emmanuelle/sphinx
      Merge pull request #1812 from blink1073/matplotlib-debug
      Disable pyamg on Travis until #1788 is closed
      Add preliminary mailmap and tool to generate it
      Re-enable pyamg, since release 3.0.2 closes #1788
      Merge pull request #1821 from jni/integral-image-fixes
      Auto-generate _RegionProps property docstrings
      Clean up style and PEP8
      Fix doc setting for Python 2.x
      Simplify cache
      radon transform: when circle=True, ignore values outside circle without throwing an error (closes #1808)
      Merge pull request #1847 from emmanuelle/remove_label_from_morphology
      Merge pull request #1866 from sergiopasra/deprecate-lena
      Merge pull request #1893 from OrkoHunter/test_thresholding_lena
      Merge pull request #1892 from OrkoHunter/test_corner_lena
      Merge pull request #1891 from OrkoHunter/test_brief_lena
      Merge pull request #1888 from ahojnnes/hessian-matrix-fix
      Merge pull request #1887 from ahojnnes/rot-fix
      draw: Add alpha ability to set_color
      Draw daisy features using anti-aliasing (closes #1665)
      Clean up implementation
      Simplify alpha vector generation
      Bump numpy requirement to 1.7.2
      Update contributor's guide: all PRs must get two +1's by core team before merge
      Revert back usage of full_like to support older numpy
      Add polygon perimiter drawing
      Update notes on algorithm.  Switched from Liang-Barsky to Sutherland-Hodgman.
      Document output of polygon_area
      Fix typos from PR feedback
      Avoid matplotlib testing problems on Travis-CI
      Add missing import
      PEP8 cleanups
      Update coordinate convention, inline matplotlib `clip_to_bbox`
      Polygon clipping requires mpl, but it is now a dependency.  Remove warning.
      Work around end-point duplication bug in Matplotlib
      With matplotlib dependency raised to 1.3, remove inlined clipping code
      Use row/column instead of x, y convention
      Up minimal six dependency to work with minimal Matplotlib
      WIP
      Raise at least one error for parallel tests.
      Extend plugins with `imshow_collection`
      Add `reset` and `compare` to novice module
      Ensure array data is always uint8
      Catch known Matplotlib warnings
      Fix assertion in novice test for older numpy versions
      Fix ORB test
      Merge pull request #1970 from alexandrejaguar/master

Steven Silvester (638):
      Initial commit of adapthist modifications.
      Tests pass and example works on build.
      Added Notes to the docstring.
      Pep8 fixes.
      Replaced generic ACM license with the Graphics Gems IV license.
      Improved support for color images using lab transform.
      Passed all tests with full coverage, PEP8 and Pylint performed.
      Translated clahe C code to python, added to list of contributors.
      Removed sys.path hack in test, fixed spelling of import in __init__.py
      Cleaned up adapthist func to use LAB for all RGB->required rest result change
      Initial commit of adapthist modifications.
      Tests pass and example works on build.
      Added Notes to the docstring.
      Pep8 fixes.
      Replaced generic ACM license with the Graphics Gems IV license.
      Improved support for color images using lab transform.
      Passed all tests with full coverage, PEP8 and Pylint performed.
      Translated clahe C code to python, added to list of contributors.
      Removed sys.path hack in test, fixed spelling of import in __init__.py
      Cleaned up adapthist func to use LAB for all RGB->required rest result change
      Rebased to master, incorporated view_as_blocks, cleaned up interpolate method and example
      Merge branch 'adapthist' of github.com:blink1073/scikits-image into adapthist
      Renamed functions to equalize_hist/adapthist, with a deprecation warning for equalize
      Used the new function names in example and test.
      Fixed comment header in test file.
      Modified test to pass with matplotlib plugin.
      Removed explicit io_plugin from test
      Changed all test to use 3 significant digits.
      Fixed last remaining sig figure problem.
      Merge pull request #1 from stefanv/adapthist_doc
      Fixed error in peak_snr, now uses float arithmetic throughout.
      Lazily import nose for testing, makes skimge load in 1/3 time.
      STY: Fixed indent on docstring.
      STY: Added newline at the end of file
      Lazily import nose for testing, makes skimge load in 1/3 time.
      Merge branch 'fast-import' of github.com:blink1073/scikits-image into fast-import
      Fixed indent on nose.run, so it will run when verbose is False
      STY: Replaced indent with original formatting, no need to change.
      BUG: Remove focus from buttons which caused unintential clicking.
      BUG: Added required QtCore import
      BUG: Not all plugins would pass the adapthist test
      BUG: Fixed exposure test to cast data on load.
      ENH: Added support for python3 in travis build
      BUG: Python 3 Travis build caught this error
      BUG: Failed exposure tests on Python3, fixed
      STY: Used float comparison in test_exposure
      STY: Change minor formatting to test travis build
      STY: Minor change to docstring to test travis build
      BUG: Adapthist error caused by divisor operator change in py3k
      BUG: Fixed another division error
      ENH: Adjusted save_to_file to create a proper composite image
      BUG: Adapthist interpolation grids could be improperly spaces
      STY: Minor change to trigger Travis build
      Fix python 2.6 restoration/unwrap error
      Move version_requirements to _shared and remove skipif from require.
      Add documentation and rename import variable
      Formatting fix.
      Formatting and docstring cleanup.
      Remove confusing dummy variable
      Update comment
      Improve import behavior for rag fallback
      Add note in travis
      Add travis_retry around the conda installs
      Make suggested changes to build update and use travis_retry for all installs.
      Fix #1108
      Fix matplotlib install, make installs verbose, add spacers
      Fix typo in spacer declaration
      Add mask option to exposure.equalize_hist with test.
      Add on_scroll behavior and allow line to change width.
      Fix failing tests
      Clear the blit background when the image changes
      Fix swirl docstring to use (row, column) nomenclature
      Remove extra space
      Remove blit background when overlay changes
      Use functools.wraps in default_fallback dectorator
      Rename (x, y) -> (row, col) in _geometric.py.
      Merge pull request #1139 from grlee77/viewer_checkbox
      Add ndim checks in filter.edges
      Catch PIL imread problems early and generate error message
      Fix indent
      Perform im.getdata check up front
      Clean up error message.
      Point user to Pillow docs for external libraries
      Minor formatting of error message
      Implement new assert_nD utility function
      Implement assert_nD in filter and feature packages
      Allow assert_nD to take multiple dimensions
      Use assert_nD in template.py
      Fix syntax error
      Fix syntax error and weiner check
      Add docstring to assert_nD
      Fix assert_nD call in template.py
      Fix canny test to reflect new error type.
      Make ndim explicit arg for clarity and update docstring
      Catch a few more asserts
      Add ndim argument in utility method
      Update build requirements and documentation
      Reorganized dependencies in DEPENDS.txt.
      Fix install_requires string formatting
      Fix python 3 unicode error
      Remove unicode decode
      Fix python3.3+ builds
      Look for  version instead of  version
      Add description for networkx optional requirment
      Merge pull request #1179 from stefanv/linetool_handle_style
      Merge pull request #1180 from stefanv/doc/widget_callback
      Merge pull request #1181 from stefanv/doc/upgrade_binary_win_install
      Make matplotlib a requirement.
      Fix travis syntax errors
      Fix another travis syntax error
      Move matplotlib to a runtime requirement
      Remove redundant 'usage requirements'
      Replace PIL with matplotlib in build_versions.py
      Fix typo in travis preventing mpl 1.0 from installing
      Install matplotlib build dependencies
      Fix matplotlib base version and install libjpeg-dev
      Install libjpeg-dev, not build-dep libjpeg-dev
      Try explicit apt installs for matplotlib
      Allow for unicode version info
      Try a newer mpl due to Tkinter err on install
      Fix error in viewer test when mpl is present but not qt.
      Avoid type incompatibility warnings
      Add networkx to requirements
      Update depends and add simpleitk and imread
      Update travis to better reflect requirements
      Allow for 3-part dependency versions
      Print full required version str
      Install networkx up front
      Do not require jpg support from io plugins in novice test
      Add pillow to requirements (but allow for PIL)
      Move Pillow to runtime reqs in DEPENDS
      Fix setup version requirement handling
      Fix __import__ fromlist
      Add the rest of the packages to the build_versions script
      Use distutils LooseVersion for version check
      Install pillow from wheelhouse
      Use strings as required by LooseVersion
      Fix syntax for python3 in setup.py
      Turn off doc examples to see if the build passes otherwise
      Fix handling of version_attr for python3
      Do not change matplotlibrc settings until after pyqt install
      Only install pyamg on py2.6 and py2.7
      Reinstate pyqt install.
      Try a reinstall of matplotlib with pyqt support
      Fix yaml syntax
      Revert to mostly working travis
      Bump matplotlib to 1.2
      Try matplotlib 1.1.1
      Fix required version
      Update to use wheelhouse
      Fix travis syntax and use tools/header.py
      Clean up travis
      Remove print statements
      Use all min reqs on Python2.7, fix syntax error and variable overshadow
      Use github downloads for specific versions on Py2.7
      Fix travis syntax errors
      Fix syntax error and QT API var handling.
      Fix string interpolation error
      Clean up required versions
      Another attempt at variable substitution
      Yet another attempt at variable substitution
      Fix installs for libfreeimage3 and astropy
      Fix install of imread
      Clean up simpleitk description and imread link
      Add notes about Travis usage
      Add notes about  and variable substitution
      Use tabs to line up if blocks
      Update Travis notes
      More updates to travis documentation
      Fix handling of non-uint8 images for PIL imsave.
      Add tifffile.py and its license.
      Add tests for new behavior.
      Remove old tifffile plugin in favor of using it in PIL plugin.
      Add tifffile.c and associated build target.
      Clean up tests and add to bento.info.
      Fix PIL tests, move helpers to _shared, fix widgets test
      Fix deprecation warning in draw.py from numpy 1.9
      Ignore tifffile.py in coveragerc
      Reinstate tifffile_plugin for backwards compatibility
      Remove temporary file and clean up imports
      Reinstate tifffile test
      Clean up imsave behavior and beef up docstring
      Add references in docstrings
      Properly handle all image types for PNG, beef up docs
      Fix handling of temp file removal
      Fix reference citation
      Do not try and save signed integer output, improved signed checks for tif.
      Fix docstring typo
      Rename the convenience methods
      Clean up pil and tif tests
      Remove unused import
      Fix relative import and test import convention
      Try removing the system numpy
      Force numpy uninstall
      Make sure there is no previous matplotlib or scipy
      Numpy is the only one installed
      Add 'imsave' to the '*' output of pil_plugin
      Update doc / trigger a travis rebuild
      Make pil the default plugin
      Fix docstring links
      Push parts of travis into separate shell scripts
      Remove variable exports
      Add shebang lines
      Add explicit sh executable
      Reorganize files
      Use an explicit matrix
      Try removing the explicit env
      Put the TEST_ARGS in travis_setup.sh
      Specify Qt API
      Rename variable to MPL_QT_API
      Put the whole script in an sh file
      Make shell scripts executable
      Remove the travis numpy
      Rename the script
      Switch shell to bash
      Use more generic shebang
      Try to get system matplotlib to pass requirments.txt
      Fix shebang line
      Update the notes
      The python path should include the parent directory now
      Fix formatting of 'for' loops
      Clean up formatting and a comment
      Keep consistent with calling convention
      Put the WHEELHOUSE as a global variable
      Put the other env variables up top as well
      Move dependency output to setup script
      Install skimage so we do not have to fool with PYTHONPATH
      Use a straight install
      Move env variable to before_install
      Add whitespace
      Remove whitespace
      Reinstate the separate build step
      Move the rest of the code to the scripts.
      Use a null backend during the doc examples
      Formatting
      Matplotlibrc was in a different location in mpl 1.1.1
      Fix MPL_DIR location
      Add a try/catch for RectangleTool.disconnect_events
      Disconnect the events manually
      Use a manual hack to remove the canvas events for old MPL
      Make sure we disconnect all the signals
      Fix comment typo
      Use the PyPI tifffile instead of bundling
      Bump the required tifffile version
      Fix tifffile imports
      Update the setup script to install tifffile
      Fix install_requires
      Remove install_requires altogether
      Use retries where applicable
      Break script into parts and travis_retry optional installs
      Make new script executable
      Export the MPL variables
      Fix handling of variables - use local
      Add tifffile to a new external package.
      Add the __init__.py to make it a package
      Remove tifffile as an external dependency
      Use new wheels site and do not install tifffile until optional
      Add tifffile.c to the build
      Fix path to tifffile.c
      Rename file to a valid module name.
      Do not test or cover the external libraries directly
      Use proper nose option
      Just disable doctests in tifffile_local
      Use tifffile 0.3.2, which passes doctests
      Use newer version of tifffile.py.
      Move tifffile license description
      Avoid star import
      Free disk space during travis install
      Quiet down optional installs to avoid hitting 10000 line limit
      Use recommended extern import method for cython
      Merge pull request #1210 from stefanv/doc/io_null_doc
      Merge pull request #1211 from stefanv/bug/setup_dependency_error
      Remove null plugin since we have the pil plugin as a requirement
      Update tests to use default plugins.
      Clean up the imports in the pil_plugin
      Use matplotlib as the default imshow
      Merge pull request #1217 from jni/morpho-out
      Merge pull request #1219 from jni/reconstruct-error
      Add tifffile support for MultiImage
      Update multi_image test and install tifffile for travis.
      Fix implementation errors
      Add tifffile support for MultiImage
      Update multi_image test and install tifffile for travis.
      Fix implementation errors
      Cleanup
      Add animated gif, a test for it, and tests without conserve_memory.
      Fix problems from rebase.
      Add a comment about reading in order
      Use pil_to_ndarray to load PIL images
      Fix implementation of pil_to_ndarray in MultiImage
      Allow multiple frames in PIL data in imread
      Allow PIL imread to read multiple images
      Add a note about sequential access.
      Change MultiImage to use new imread.
      Avoid circular import
      Fix failing tests
      Finish merging multiimage capability into collections
      Pep8 cleanup
      Allow keyword arguments to imread
      Clean up docstring
      Add tests for ImageCollection with multi_images
      Close the image file pointer when done.
      Test collection slicing with multi-images.
      Indentation fix
      Merge pull request #1046 from stefanv/osx_wheels
      Merge pull request #1184 from stefanv/filter_rename
      Move debug info into top level travis for visibility
      Merge pull request #1227 from stefanv/remove_build_status
      Merge pull request #1232 from jni/negative-histogram
      Merge pull request #1240 from ahojnnes/edge-sign
      Use virtualenv for all builds and install everything from wheelhouse
      Use system packages on Py2.7
      Clean up use of python version check for pyqt install
      Add install_requires for six
      Fix python version check
      Invoke sudo to create link to system PyQt
      Merge pull request #1262 from stefanv/gitter_travis
      Move the virtual framebuffer to top level yaml
      Fix handling of imgnum
      Fix handling of multi-image gifs and add test
      Reinstate run_module_suite in test_pil
      Fix file indexing in ImageCollection
      Fix handling of singleton frames in pil_to_ndarray
      Test all modules (including underscores), and always doctest
      Use the test arguments from skimge.test_verbose
      Update test args for command line
      Do not use importlib in tifffile_local to avoid warnings
      Improved docstring
      Beef up docstring
      Increase timeout to install requirements
      Set the pip timeout globally for Travis
      Install simpleitk from wheelhouse if available
      Merge pull request #1274 from blink1073/use-wheelhouse-simpleitk
      Do not try and install SimpleITK on Py3.2 or 3.4
      Fix boolean logic for SimpleITK install and add warning
      Fix typo in bash conditional
      Create a new figure for imshow if there is already data
      Prevent travis-try breaks when recreating hard link for PyQt
      Add python3 doc building capability
      Add a script to create env variables and travis folding helpers
      Switch to a single script to get the folds to work
      Use our own retry command since travis_retry is not exported
      Use the system python for 2.7
      Do not explicitly remove packages from requirements.txt
      Fix usage of set command
      Fix section titles to avoid spaces
      Add a coveralls repo token
      Update travis_notes and move to tools folder
      Reinstate the pip timeout
      Shuffle some commands back to before_install
      Merge pull request #1284 from emmanuelle/fix_dox
      # This is a combination of 7 commits.
      Handle warnings in several packages
      Handle more warnings and reset io plugins as needed
      Accommodate upcoming changes to RectangleSelector API
      Clean up PIL plugin and handle more warnings
      Fix some doctest warnings
      Import scipy before turning on warnings
      Add a test setup helper function
      More doctest and pil_plugin fixes
      Handle more warnings
      Create new expected_warnings helper and some cleanup
      More cleanup and updating of tests
      Add a random seed to the setup
      Fix handling of multiple warnings and update tests
      Allow for no exceptions to be raised and cleanup
      Fix label wrapper function
      Change default neighbors to None
      Import scipy.sparse in setup_test
      Allow warnings in the minimum build
      Fix Py2.7 build errors
      Fix syntax error and missing | operator
      Fix final failing test on Py2.6 build
      Update docstring and fix pyamg warning check
      Fix scipy.sparse.sparsetools warning
      Fix indent in docstring example
      Style fixes
      Draw the canvas when creating the ax to avoid RuntimeError
      Try using matplotlib cleanup decorator
      Remove unnecessary measure.label shim
      Update viewer for PyQt5 compatibility
      Use a smarter backend selection
      Do not try and guess the mpl backend
      Remove debug raise
      Allow viewer package to import without Qt
      Do not use ImportError behaviour in viewer
      Fix import error in viewer.utils.core
      Fix another viewer import
      Fix the dummy Signal
      Another fix for dummy Signal
      Instantiate QtCore dummy class
      Add dummy Qt properties
      Add QWidget to the mock Qt
      Fix mock Qt objects again
      Merge pull request #1314 from SalvatoreScaramuzzino/RectTool_docs
      Fix lingering merge issue
      Merge pull request #2 from flying-sheep/pyqt5
      Separate qt version check from has_qt flag
      Merge pull request #1328 from pratapvardhan/misc-doc
      Merge pull request #1300 from jni/io-imshow-data
      Update docstring and create test module setup function
      Fix filter package shim implementation
      Merge pull request #1341 from JDWarner/cover_pad
      Merge pull request #1324 from ahojnnes/rel-import
      Force novice.Picture.from_size to return a uint8 image
      Add handling of CMYK images
      Avoid creating an int64 array
      Merge pull request #1361 from sciunto/doc2
      Add test for cmyk and use relative imports
      Reinstate run_module_suite
      Do not add alpha channel to CMYK images
      Compare the images in LAB space
      Merge pull request #1365 from jni/future
      Merge pull request #1366 from JDWarner/mixed_permission_error
      Cannot use bare deprecated decorator
      Give an alternate canny location
      Add an explicit test for canny deprecation
      Fix canny import test
      Merge pull request #1370 from r3vl1s/imread_collections
      Update test to reflect new error behaviour
      Fix some sphinx warnings and add to build
      Skip build on 3.2
      Remove imshow from PIL plugin
      Update the README with some warning hints
      Install latex and dvipng
      Do not build docs on Py2.6
      Be more specific about citation not found error
      Remove citations in first line of docstring
      Fix links
      Remove unnecessary option and update test
      Fix behavior of skimage.doctest when nose is not found
      Test against PIL on python 2.7
      Fix name of pil package
      Merge pull request #1382 from arve0/ImageViewer.update_image
      Merge pull request #1384 from stefanv/py3_ghpages
      Merge pull request #1376 from jmetz/threshold_li
      Merge pull request #1392 from jmetz/threshold_li
      Merge pull request #1397 from JDWarner/fix_find_contours
      Merge pull request #1406 from anntzer/plugins-relative-imports
      Add a helper function to check for low contrast
      Handle Sphinx 1.3 API change
      Try a fix for py27 build
      Fix support for Sphinx 1.3
      Revert changes to force Sphinx 1.2.3
      Remove unnecessary sphinx version check
      Merge pull request #1428 from arve0/doc_requires_latex
      Merge pull request #1429 from JDWarner/dev_guide_indentation
      Merge pull request #1416 from jni/low-contrast-docstring
      Merge pull request #1430 from arve0/document_measure.label
      Merge pull request #1436 from ericdill/save-buttons-fix
      Fix handling of ImageCollection indexing and add test
      Add a note to the docstring
      Remove debug error raise
      Always use _find_images()
      More specific test in freeimage
      Do not attempt to use _frame_index for exotic load_patterns
      Fix freeimage collection test
      Merge pull request #1426 from arve0/dependencies
      Fix for broken setuptools
      Fix setup to allow installing from PyPI
      Another random warning guard
      Only index tif and gif files in collection
      Fix handling of non-gif and tiff files
      Do not read in the whole array
      Merge pull request #1460 from jni/fix-pil-imsave
      Merge pull request #1464 from jni/fix-doc
      Merge pull request #1467 from stefanv/doc/novice_examples
      Merge pull request #1469 from stefanv/doc/gh_pages_version
      Merge pull request #1486 from emmanuelle/remove_image_class
      Merge pull request #1487 from hawkerpl/pillow-imread-error-message-fix
      Merge pull request #1492 from arve0/viewer_button
      Merge pull request #1498 from cgohlke/patch-2
      Add a multi-page rbg tiff file
      Fix handling of img_num=0 cases
      Add tests for multipage_rgb tif
      Move tests to pil plugin and fix file path
      Fix failing pil test
      Remove explicit call to use_plugin
      Reinstate explicit plugin
      Add a multipage tiff test and cleanup test_tifffile
      Merge pull request #1501 from stefanv/doc_fixes
      Merge pull request #1499 from stefanv/doc_style
      Merge pull request #1532 from JDWarner/fix_appveyor
      Allow view_as_window to take a tuple step, and update tests
      Fix typo
      Merge pull request #1545 from cgohlke/patch-1
      Merge pull request #1543 from stefanv/docs/remove_logo
      Style and doc updates
      Merge pull request #1549 from JDWarner/fix_tiffile_c_replacement
      Create an optimal step parameter for view_as_windows
      Ensure consistent handling of division
      Use files from tifffile 0.6.2 package
      Allow plot plugin to have tools
      Merge pull request #1594 from stefanv/doc/remove_toc
      Merge pull request #1590 from jjhelmus/faster_tests
      Merge pull request #1518 from stefanv/relative_imports
      Mimic the way scipy handles numpy in setup.py
      Fix typo
      Fix docstring
      Use a container and pip cache
      Do not let PIL handle the file handle
      Remove unclosed file guards
      Fix imsave test
      Remove offending line
      Allow pil imread to accept file-like objects again
      Fix variable name
      Try fixing the remaining unclosed error
      Use context manager for tif imread as well
      Fix handling of tiffile handles
      Move the tifffile wrapper to tiff plugin
      Fix tifffile plugin import
      # This is a combination of 10 commits.
      Update equalize_adapthist to use new view_as_windows
      Style fixes
      Update docstring
      Docstring and TODO
      Update tests
      wip
      Fix rebase error
      Rename optimal_step and add comments
      Fix view as windows test
      Remove min_overlap kwarg
      Use smarter indexing
      Remove debug print
      Rename variable for clarity
      More variable renaming
      Remove use of view_as_window and fix docstring
      Fix failing test
      Add the viridis colormap and make it the default
      Always create a colorbar and update tests
      Clean up tests by using fresh figures
      Add another file guard
      Make all of the pip installs quiet to try and fit in 10000 lines
      Try only warning once per module
      Use default warning print behaviour
      Enable deprecation warnings
      Try and install pyqt on the container
      Add an explicit pip cache since we override the install step
      Try using debug and all skimage warnings
      Be more explicit about configparser import
      Do not try and update scipy on a pip install if already installed
      Do not try and upgrade larger dependencies on skimage install
      Use a less-invasive approach
      Allow viridis to be used by name
      Fix relative import
      Only add colorbar if the cmap is not the default
      Merge pull request #1639 from stefanv/gray2rgb_alpha
      Merge pull request #1688 from robintw/add-make-bat-for-docs
      Merge pull request #1687 from connellybarnes/master
      Merge pull request #1694 from ahojnnes/regionprops-dtype
      Merge pull request #1683 from jwiggins/travis-tweaks
      Add a plugin and test for imageio
      Update the dev notes
      Remove unnecessary import
      Add imageio to optional reqs
      Do not try optional dependencies on Py3.2
      More cleanup of installation
      Fix travis script syntax
      Merge pull request #1681 from emmanuelle/fix_pep8_gallery
      Add dev docs push on master builds
      Add the deploy_docs script
      Fix document build error with new notebook package
      Allow polygon to take a list or tuple and add a test
      Fix doc build and add instructions
      Merge pull request #1699 from jwiggins/use-ccache
      Merge pull request #1708 from JDWarner/fix_win_doc_build
      Add api docs and gallery on Windows build and build docs on Appveyor
      Build from the right directory and add api to latex command
      Add the make target
      Fix PYTHON executable and html script
      Add a 64bit python 3.4 build
      Use the latest pip on appveyor
      Fix pip upgrade
      Try a 64bit 2.7 build
      Fix yml format
      Revert to single build
      Merge pull request #1710 from sciunto/favicon
      Fix variable check
      gh-pages is the default target branch
      Update token
      Try with debug output
      Merge pull request #1717 from sciunto/doctravis
      Encrypt token for target repo
      Fix variable name
      Another fix
      Fix variable name
      Try with a token encrypted in this repo
      Add notes on how to deploy docs on Travis
      Merge pull request #1724 from sciunto/unclosedfile
      Test against Python 3.5 on Travis
      Use Dask 0.7.1 on Python 3.2
      Remove 3.2 from travis build
      Move tiff-file handling to top level
      Pop the img_num keyword if given
      Remove the tifffile save tests from test_pil
      Remove another tiff check in test_pil
      Fix adapthist mask logic
      Merge pull request #1734 from sumitbinnani/RenamingFilter
      Merge pull request #1756 from stefanv/doc_with_gallery
      Prevent infinite loop in adapthist for low clip_limit
      Fix setup.py in the presence of no numpy
      Fix errant docstring
      Fix matplotlib backend
      Remove debug print
      Allow 64bit integer conversions and direct downcast
      Update failing tests
      Add debug print
      Fix failing test
      Improve warning message
      Move the doc build readme to contributing.txt
      Start deprecation of python 2.6
      Merge pull request #1691 from juliusbierk/active-contour-model
      Merge pull request #1824 from stefanv/travis_pyamg
      Merge pull request #1823 from emmanuelle/structure-gallery
      Merge pull request #1834 from emmanuelle/active_contour_example
      Merge pull request #1603 from michaelpacer/draw_coords
      Lower required matplotlib version on 2.6 test
      Merge pull request #1848 from soupault/doc_edge_modes
      Allow axes to be provided to imshow and fix colorbar
      Remove debug message
      Merge pull request #1849 from emmanuelle/fix_contribs
      Use a stacklevel of 2 by default for all warnings
      Include the pyx files
      Make warn() more explicit
      Cleanup the _warnings module
      Work around limitation in warnings.warn
      Cleanup
      More cleanup
      Add missing space
      Debugging appveyor warning
      Add a note about the new warning
      Add another note about the new warning
      Merge pull request #1919 from ahojnnes/rgb2gray
      Merge pull request #1918 from ahojnnes/examples
      Merge pull request #1930 from emmanuelle/release_notes
      Merge pull request #1927 from sciunto/doc_inpaint2
      Merge pull request #1905 from stefanv/novice_undo
      Merge pull request #1937 from stefanv/test_parallel_fail

Stuart Mumford (1):
      Website font readability improvements.

Tabish (11):
      Modify 'integral_image' to support nD
      Make the nD integral_image backward compatible
      Changed integral_image parameters to new format
      Updated test_integral to support new format of parameters
      Corrected documentation/indentation, changed vstacking
      Corrected Python3 slice indexing error
      Check input for Iterables instead of numeric types, PEP8,indentation,added comments
      Corrected doctest output
      Corrected grammatical mistakes in documentation
      Removed extra Tab character
      Changed input format,updated docstring, PEP8

Thomas Robitaille (4):
      Use ``cythonize`` instead of hard-coding Cython commands. Fixes #829.
      Only require Cython if using a developer version
      Use Astropy by default if possible
      Added entry for Astropy in DEPENDS.txt

Thouis (Ray) Jones (13):
      Constant time per pixel median filter
      Fix docstring to match new default value and constraint for radius.
      use SSE to add/subtract histograms
      Rewrite histogram operations to remove GPL license on Perreault's code
      fix UTF-8
      remove typedef, use numpy type in its place
      add casts to suppress compiler warnings
      move _histogram.h to _shared/vectorized_ops.h, and copyright info to CONTRIBUTORS.txt
      pep8
      remove _histogram.h from setup.py, it's been moved to _shared
      BUG: not all histogram memory was aligned correctly
      fix rank filters for float structuring element
      Merge pull request #1 from ahojnnes/thouis-master

Tim Seifert (2):
      load correct _frame_index
      fix for concatenate test

Tim Sheerman-Chase (18):
      Initial work to introduce piecewise affine transform
      Achieved functional piecewise affine transform
      Update doc strings with piecewise affine
      Improve comments
      Improve comments
      Setting _matrix to None, because the transform is not of this form
      Wrap long line
      Initial attempt at optimising HOG features with cython
      Fix comments and tab whitespace
      change import syntax for python 3
      Add _hoghistogram.pyx to bento.info
      Merge pull request #2 from ClinicalGraphics/fasthog
      Merge pull request #3 from ClinicalGraphics/fasthog
      Merge pull request #4 from ClinicalGraphics/fasthog
      Merge pull request #5 from ClinicalGraphics/fasthog
      Merge pull request #6 from ClinicalGraphics/fasthog
      Merge pull request #7 from ClinicalGraphics/fasthog
      Merge pull request #8 from ClinicalGraphics/fasthog

Tomas Kazmar (2):
      BUG: Fix Orientation for diagonal regions in regionprops.
      Wrap long lines.

Tony (5):
      Add palette images
      Fix imread so palette images are converted to grayscale or RGB images
      Add test for palette_is_grayscale
      Add list of build requirements
      Rewrite as_grey property without enhanced property decorators to be python 2.5 compliant

Tony S Yu (699):
      Remove print statements
      Fix order of keyword arguments to _tv_denoise_2d.
      Make test_tv_denoise_2d more strict.
      Reorder eps and weight keyword args to tv_denoise
      Use matplotlib's plot_directive.py for Sphinx documentation.
      Fix import in apigen.py, which fails when other scikits are present.
      Change Makefile so that autogenerated API docs are removed with "make clean".
      Replace plot_directive.py with copy from matplotlib's trunk.
      Fix lengths of restructuredtext section headers in docs.
      Fix docstring so Sphinx doesn't complain about indentation.
      Remove index.html template and use markup in index.txt instead.
      Add version number to index page of docs
      Format front page so it looks similar to old template.
      Initial commit with working logo generator
      DOC: Fix order of parameters passed to cp
      DOC: adjust default table styling for a cleaner look
      DOC: make index.html the master doc
      DOC: fix copyright
      Add parameter to flip structuring element about its origin
      Remove unnecessary if-block.
      Reuse open/close operations for white/black tophat
      Add tests for correct behavior of square structuring elements
      Fix morphological open/close for square structuring elements with even sides.
      Move test functions to test class
      Add tests for rectangular structuring elements
      Replace flip parameter in erode/dilate with shift_x and shift_y
      Minor fix to docstring
      Improve docstrings in grey morphology module
      Add examples to grey morphology docstrings
      Specify that input arrays to grey morphology filters must be uint8
      Reformat doctests so that tests pass
      Require median filter radius >= 2
      Switch radius and mask arguments for median_filter
      DOC: Add note about changed arguments for median_filter
      Merge pull request #81 from NeilYager/neil_yager-io_fix
      Merge pull request #85 from cgohlke/patch-2
      Add thresholding module with Otsu's method to calculate threshold.
      Rename otsu_threshold to threshold_otsu
      Add handling for images with negative integers
      Change docstring threshold_otsu output
      Remove binarize function
      Add thresholding example
      Remove thresholding example from user guide in favor of docstring example.
      Fix minor typo in docstring example
      Remove thresholding subpackage and integrate into filter subpackage.
      Merge pull request #91 from stefanv/imshow_plugin_scales
      Minor changes in wording.
      Rename `bins` parameter to `nbins`.
      Merge pull request #101 from stefanv/pil_bilevel_png
      Fix imread regression for int16 images with PIL plugin.
      Add test for reading uint16 image
      Add test using big-endian uint16 image.
      Add example for threshoding
      Improve thresholding example
      DOC: emphasize annotation in histogram.
      Merge pull request #98 from tonysyu/thresholding
      Add exposure module with histogram equalization function
      Change API for equalize_hist and cumulative_distribution.
      Rename equalize_hist to equalize and minor cleanup
      Remove `density` parameter in `histogram`
      Add rescale_intensity function with test and example.
      Merge pull request #108 from WarrenWeckesser/ph-doc-fix
      Merge pull request #106 from NelleV/harris_corner_detection
      Minor modifications to Harris corner detector
      Fix peak detection algorithm.
      Refactor peak detection algorithm from Harris detector.
      Make test more robust.
      Use ndimage.maximum_filter to simplify peak detection.
      Rename peak_min_dist to peak_local_max.
      Move harris corner detection from filter to feature subpackage.
      Fix casting error when converting from int8 to uint16.
      ENH: Add tests for full coverage.
      ENH: Check that float images are between 0 and 1.
      Merge pull request #1 from stefanv/tony_rescale_intensity
      Merge pull request #105 from stefanv/label_cleanup
      DOC: Change body layout so that it doesn't wrap around sidebar.
      DOC: Adjust font sizes.
      DOC: Remove duplicate css definition.
      DOC: Increase spacing between paragraphs.
      DOC: Adjust heading size and color.
      DOC: Adjust header and link colors.
      DOC: Remove duplicate "pre" formatting.
      DOC: Adjust link font in contentstable.
      DOC: Tweak layout of contents table.
      DOC: Adjust spacing between sections of sidebar.
      Fix test: clip image that exceeded float range.
      DOC: Adjust size of tt text (i.e. ``text``).
      DOC: Adjust spacing after `pre` block.
      DOC: Fix grid layout of gallery.
      DOC: Improve description of image data types in user's guide.
      Merge pull request #116 from stefanv/matplotlib_interpolation
      Merge pull request #120 from stefanv/scivi_rename
      DOC: Add description of subpackages to package docstring.
      DOC: Add sidebar section listing package versions.
      Minor fixes to build process.
      DOC: Use CSS to style versions sidebar.
      DOC: Add (u)int32 data type ranges to table.
      DOC: Tweak superscript formatting.
      DOC: Add example of bad dtype conversion.
      Merge pull request #124 from stefanv/doc-gallery-show-version
      Merge pull request #130 from stefanv/grey2rgb
      DOC: Convert README to markdown for proper github rendering.
      Merge pull request #136 from stefanv/plugin_consistency
      Clarify difference between view_as_blocks and view_as_windows.
      DOC: Limit figures to 8 inch widths.
      DOC: minor example clean up.
      DOC: tweak image sizes in gallery.
      DOC: Add note about figure width to developer guide.
      Merge pull request #146 from stefanv/imread_default_plugin
      DOC: Add hidden toctree for main index.
      DOC: Add toctree to example gallery.
      DOC: Add hidden toctree to tasks
      DOC: Remove completed tasks.
      DOC: Delete completed task.
      DOC: Remove "completed" labeled-image plotting task.
      Merge pull request #153 from cgohlke/patch-2
      Merge pull request #154 from cgohlke/patch-3
      Fix test when tifffile plugin is not available.
      FIX: Check plugin is loaded before checking its functions.
      Merge pull request #159 from 'stefanv/version_remove'
      Remove version.py reference in Makefile
      Merge pull request #163 from stefanv/convert_with_copy
      Fix import
      Deprecate threshold and replace absolute and relative thresholds.
      Change threshold test from >= to >.
      Fix: return empty list for flat image.
      Add docstring comment about flat peaks.
      Rename for clarity
      ENH: Add `num_peaks` to limit the number of peaks.
      Merge pull request #166 from 'vincent-albufera/exemple_modif'
      BUG: fix logging for Python < 2.7.
      DOC: Add example of adaptive thresholding.
      Change "text" image to grayscale.
      Add "page.png" and use for threshold example.
      Merge pull request #185 from stefanv/doc_math_css
      Remove unused imports
      Rename greyscale morphology functions.
      Make test module runnable.
      Fix dtype compatibility for functions in morphology.grey
      Replace `import *` with `import grey`.
      Remove unused variable.
      Add `__all__` to grey module.
      Rename test module to match module.
      Merge pull request #184 from stefanv/fast_coordinate_map
      Convert template matching Pull Request to skimage
      Remove unused imports
      Fix whitespace
      Update setup file to new repo and fix typos
      Remove dependency on math module
      Wrap long lines
      Clean up docstrings
      Fix copy-paste bug
      Wrap long lines
      Change OpenCV import to make it an optional dependency
      PEP8: add spacing around operators
      Remove unused `integral_image` function in _template.pyx
      Simplify _template.pyx using integral_image from transform subpackage.
      Add comment about code duplication.
      Fix assert failure in example
      Replace assert statement with plot to show matches
      Reorganize example so that all plotting code is at the end
      Move template matching to feature subpackage
      Change _template.match_template variables to float.
      Fix type errors in `match_template` by casting inputs to float32.
      Use `feature.peak_local_max` instead of custom peak detection.
      Remove OpenCV version of match_template
      Refactor template matching.
      Remove unnecessary truncation at boundary.
      Add `pad_output` argmument to `match_template`.
      DOC: demonstrate where the template is matched in an image.
      Simply equation in docstring of `match_docstring`.
      Fix bug when indexing into summed-area table.
      Minor cleanup
      Rewrite normalization algorithm.
      Tweak example to reduce confusion.
      Add normalization test and remove unnecessary clipping.
      Renaming for clarity.
      Fix convolution input to get correlation results.
      Pad input image instead of output.
      Change `flipud`/`fliplr` to array indexing.
      Add alternate example for `match_template`.
      Check that image is larger than template.
      Prevent `match_template` from returning NaNs.
      Set `match_template` to default to no padding and fix test.
      Rename `pad_output` parameter to `pad_input`.
      Add test for `pad_input = True`.
      Add doctest example to `match_template`.
      Fix template example.
      Fix shape unpacking ((height, width), not (w, h)).
      Add new example plot for `match_template`.
      DOC: Replace template example with alternate example.
      Add contribution note.
      BUG: fix test images (float images between 0 and 1).
      DOC: Remove Sphinx extension for inheritence trees.
      DOC: Replace use of gen_rst with plot2rst extension.
      BUG: Fix stackcopy for grayscale image.
      DOC: convert segmentation example to tutorial-style example.
      DOC: Set rcparams for better default plots.
      Remove gen_rst
      Fix: check all io functions when loading of default plugins.
      Merge pull request #194 from jehturner/pyfits_hdusize_fix
      DOC: Fix link formatting
      Refactor how PIL is set as default for imread.
      Replace mpltools reference with skimage.
      Merge pull request #119 from stefanv/ssim
      Merge pull request #208 from stefanv/io_reset
      Skip test that fails for PIL < 1.1.7
      Add morphological reconstruction with test and example.
      DOC: Add list of tasks for SciPy 2012 sprints
      ENH: Add orientation kwarg to IntelligentSlider
      ENH: Add image viewer based on Qt and Matplotlib
      API change: switch order of image viewer and callback arguments.
      Add PlotPlugin and cleanup code.
      Add LineProfile plugin.
      Fix: Move on_draw method to base Plugin
      Merge pull request #227 from thewtex/thresholding_failures_2
      Clean up old code and add docstrings.
      Delete overlay when deleting plugin.
      Add docstring for `overlay` and reorder methods.
      Show coordinate and intensity info in status bar.
      Add plugin names
      Add `update_on` parameter to slider and allow update on release.
      Refactor image overlays to special plugin base class.
      Move OverlayPlugin and PlotPlugin to their own modules.
      Change ImageViewer to automatically call plugins.
      ENH: Align image and plugin windows
      Fix `add_argument`.
      ENH: Simplify creation of Slider widget.
      ENH: Rename callback functions for clarity.
      ENH: Simplify widget addition.
      ENH: Move closeEvent definition to base class.
      ENH: Generalize add_widget function.
      ENH: Allow `Plugin.add_widget` to hook into Plugin attributes.
      Make alpha value to ClearColormap adjustable.
      API Change: Attach ImageViewer to Plugin after init.
      Minor cleanup.
      Merge pull request #226 from jni/jni-alphanumeric-sort
      Change image_viewer to Plugin property.
      ENH: Let Qt handle most of the window sizing.
      Rename (dis)connect_event to (dis)connect_image_events.
      DOC: Add todo note
      ENH: Display overlay by default
      ENH: filter image when Plugin is attached to ImageViewer
      BUG: in-place add should return object
      ENH: Add example of adding widgets to plugin
      ENH: Change inplace-add to normal add to support alternate syntax
      BUG: Don't override `image_filter` method if defined by subclass
      Merge pull request #234 from bdholt1/fix-hog
      STY: Remove unused `add_artist` method.
      DOC: Improve docstrings for ImageViewer.
      DOC: Improve docstrings for Plugin class.
      ENH: allow color to be set by name
      STY: reorder methods for clarity.
      DOC: clean up docstring for PlotPlugin
      DOC: cleanup docstring and reuse parameter defined by parent class
      DOC: Add class docstring
      STY: Refactor MatplotlibCanvas from ImageCanvas and PlotCanvas.
      ENH: Create new Slider with editbox.
      Revert modifications of IntelligentSlider.
      Remove lineprofile temporarily (saved in a branch)
      Remove unused import
      STY: Refactor BaseWidget from Slider and ComboBox
      BUG: Fix behavior when initial overlay limits are bad.
      DOC: Explain use of callback parameter.
      DOC: Add explanation of add operator.
      BUG: Fix scaling when setting default slider value.
      ENH: Add Slider `value_type` to allow int values.
      Merge pull request #230 from dharhas/dharhas-lab-color-space
      BUG: Rename modules with duplicate function names.
      BUG: more module renaming to prevent duplicates.
      ENH: Add RequiredAttr to raise warnings when attr not set.
      DOC: Clean up docstring for Slider
      ENH: Add SaveButtons widget.
      BUG: reset image when plugin is closed.
      ENH: Add OK/Cancel buttons
      STY: Tweak button sizes.
      Merge pull request #238 from ahojnnes/regionprops-orientation
      Merge pull request #221 from jjhelmus/imagecollection_slicing
      DOC: Add note about ImageCollection slicing.
      BUG: fix import of Cython extension.
      STY: PEP8 and other clean up.
      Update function names that were changed since original PR.
      STY: minor PEP8 change
      BUG: Rename files with similar function, python module, and cython extension names.
      STY: Use standard skimage data type conversion.
      DOC: add comments to clarify algorithm
      STY: Rename returned image to distinguish input from output.
      ENH: Add reconstruction by erosion.
      DOC: Clarify code comments and docstring
      STY: Rename variables for clarity.
      DOC: Reorder docstring sections.
      ENH: Add examples of morphological reconstruction.
      DOC: Remove peak detection tutorial.
      ENH: Add test of reconstruction by erosion.
      ENH: Use Cython data types instead of Numpy dtypes.
      DOC: Fix docstring to match rank order.
      DOC: Combine examples for finding spots and filling holes.
      ENH: Add regional maxima example
      Merge pull request #254 from jaberg/default_AffineTransform
      ENH: Allow imshow to show images when given file name.
      BUG: Update filter when edit-box is changed.
      Merge pull request #255 from stefanv/coding_conventions
      Merge pull request #233 from wilsaj:image-html-repr
      Remove custom dtype range
      BUG: Closing ImageViewer shouldn't quit parent program
      BUG: Adjust pixel limits based on image data type
      DOC: Improve description of reconstruction
      DOC: Bump up Cython version up to 0.16
      BUG: Fix nosetest and autodoc errors when PyQt4 not available
      Change error to warning so that autodoc doesn't fail.
      Change reconstruction to support Cython 0.15.
      BUG: fix import of rank_order
      BUG: Fix division error for Python 3
      ENH: Change assert statement for better error output
      BUG: Bento version must end with number
      ENH: Add script to check that bento.info is up-to-date.
      BUG: Update bento.info to match setup.py files
      ENH: Add CollectionViewer
      Add line profile plugin back in.
      DOC: Fix broken doc build due to broken import
      DOC: Turn off numpydoc flag to silence warnings
      Change import to silence import errors in docs.
      DOC: Separate config sections in conf.py
      Merge pull request #292 from ahojnnes/geometric-coeffs
      DOC: Remove inheritance-diagram Sphinx role.
      DOC: Remove inherited-members Sphinx role.
      DOC: Remove commented out code.
      DOC: Fix Sphinx warnings for regionprops
      DOC: Change title to match numpy docstring standard
      DOC: Fix Sphinx warning.
      DOC: Fix formatting to prevent Sphinx warnings
      DOC: Reformat shape argument
      DOC: Fix citation syntax
      DOC: Shorten plugin description so it fits in table
      DOC: Fix section titles to match numpy doc standard
      DOC: Fix rendering bug.
      DOC: Wrap long lines
      DOC: Integrate uncompleted Scipy2012 Sprint tasks
      STY: Clean up whitespace
      Merge pull request #279 from emmanuelle/fix_doc
      STY: Clean up imports
      STY: Rename tests.
      TST: Add tests for masked region
      DOC: Add note about expanded masking
      Refactor masking
      STY: Rework masking based on suggestions by @ahojnnes
      STY: Refactor initialization of QApplication.
      BUG: Initialize QApplication when creating Plugin.
      DOC: Add quickstart for viewer to User Guide
      ENH: Set viewer min-intensity to zero when possible.
      STY: Clean up formatting and remove repeated imports
      Merge pull request #304 from ahojnnes/contours-array-fix
      Merge pull request #302 from ahojnnes/pyramids
      DOC: Add note about negative dtypes
      DOC: float dtype changed from (0, 1) to (-1, 1).
      DOC: suppress plot display in doctests
      DOC: Fix block-formatting in doctest
      DOC: Fix doctests in io collection module
      DOC: Fix array formatting in doctest
      DOC: Fix doctests in viewer subpackage
      DOC: Add missing import in doctest
      DOC: Fix doctests in lpi_filter
      DOC: rewrap docstring lines and remove unused import
      DOC: fix numpy dtype print out.
      DOC: Replace doctest with literal code block
      DOC: Remove plots from docstring examples
      Merge pull request #311 from stefanv/dtype_bool_fast
      DOC: Fix docstring note about conversion range
      Add `assert_greater` compatibility function.
      Merge pull request #320 from ahojnnes/pyramids-fix
      DOC: Fix attribution for testing function
      Merge pull request #325 from JDWarner/img_as_float_docfix
      DOC: Add 0.7 release notes
      PKG: Update version to 0.7.0
      PKG: Update docversions.js to 0.7
      PKG: Use correct version in docversions.js
      PKG: Start 0.8 development cycle.
      PKG: Clarify instructions for gh-pages.
      PKG: Update release notes.
      Merge branch 'master' of github.com:scikits-image/scikits-image
      PKG: Add attributions to contributors.txt
      DOC: Change algorithm for depth of API generation
      DOC: Format gallery into a grid
      DOC: Save thumbnails with a fixed image shape
      DOC: Remove task list from docs
      BUG: Pass package directory as positional argument to nose.run
      ENH: Add grayscale support and color adjustment to visualize_boundaries
      STY: Rename visualize_boundaries to highlight_boundaries
      Merge pull request #346 from ahojnnes/warp-zoom
      STY: Rename highlight_boundaries to mark_boundaries
      DOC: Add imports to doctests to fix errors
      DOC: Add imports to doctests to fix errors
      Merge pull request #351 from ahojnnes/test-coverage
      Merge pull request #367 from stefanv/gray2rgb_for_rgb
      ENH: Allow imread to read URLs
      Fix handling of temporary file.
      Merge pull request #358 from stefanv/userguide_dtype_range
      ENH: Add test and support for reading file URLs
      STY: Fix confusing use of context-manager.
      Merge pull request #370 from stefanv/clock
      BUG: Adapt to whitespace in text search
      ENH: Compare full paths of cython extensions.
      ENH: Add filter.rank extensions to bento.info
      ENH: Add canvastools subpckages with base class and line tools
      BUG: Move callback connection to parent class
      STY: Reuse ThickLineTool in LineProfile
      BUG: Fix blitting behavior in canvastools
      STY: Refactor parts of LineTool to CanvasToolBase
      BUG: Clean up imports
      Merge pull request #361 from ahojnnes/hough-peaks
      ENH: Make black the default boundary color for mark_boundaries.
      Merge pull request #375 from ahojnnes/hough-peak-example
      Merge pull request #383 from stefanv/windowserror
      Merge pull request #386 from ahojnnes/dependecy-check
      Merge pull request #387 from ahojnnes/peak-det
      Fix figure canvas creation.
      BUG: Allow slider to accept key press events
      ENH: Default to blitting
      BUG: Fix background update when canvas resized
      STY: Remove commented out code
      BUG: Fix file separators for windows
      BUG: Fix blitting behavior
      STY: Clean up unused blitting-related code
      STY: Minor code clean up
      ENH: Draw new line when click is far from handle
      STY: Remove reset functionality in line tool
      ENH: Increase default selection threshold for ease of use
      STY: Rename line_tool module to linetool
      ENH: Add measure tool plugin
      STY: Refactors general parts of line tool to base tool
      ENH: Add rectangle tool and crop plugin example
      ENH: Overhaul callback system.
      BUG: Fix image viewer for RGB images
      BUG: Fix file URI in test (2nd attempt)
      BUG: Fix matplotlib import for compatibility
      STY: Switch order of callback parameters
      BUG: Fix line visibility
      STY: Rename parameters for clarity
      BUG: Fix renamed parameter
      Fix typo
      ENH: Add color histogram plugin
      ENH: Handle redraw edge case.
      Remove x, y parameters from line tool to simplify API
      STY: Rename end_pts to end_points
      STY: Remove linewidth parameter to simplify API
      DOC: Add parameters to docstring
      BUG: Fix LineTool example after change to call signature
      BUG: Fix line thickness change call back
      STY: Simplify imports
      ENH: Add setter for extents property
      ENH: Improve end_points setter
      BUG: Update measure for change in line_tool API
      BUG: Fix error when matplotlib is not available
      BUG: Fix MRO bug when matplotlib not installed
      BUG: fix end_points setter when given a tuple
      BUG: Validate mouse interaction
      ENH: Allow Save buttons to save overlay
      BUG: Move import inside deprecated function
      ENH: Let CannyPlugin adapt to input image
      Merge pull request #2 from blink1073/canvastools
      ENH: Warn if using OverlayPlugin w/ matplotlib < 1.2
      Merge pull request #377 from jni/jni-segmentation-join
      Merge pull request #378 from sergeyk/master
      BUG: Fix save to stack so it saves filtered image
      Merge pull request #399 from mkcor/doc_first-steps-contrib
      ENH: Add open and save menu actions
      Merge pull request #403 from blink1073/test-failure
      BUG: Fix clean up of canvastools
      ENH: Remove unnecessary clean up functions.
      ENH: Add PaintTool and LabelPlugin
      ENH: Add demo of LabelPlugin
      Use print function for python 3 compatibility
      Merge pull request #3 from blink1073/canvastools
      ENH: Cast arrays for numpy compatibility.
      BUG: Update bento build to match setup files.
      ENH: Improve usability of watershed_demo.
      Fix alpha handling for image saving.
      Add save buttons to viewer example.
      Fix return value of deprecated `equalize` function.
      ENH: If example fails, print traceback and carry on.
      DOC: Fix Sphinx errors.
      Change `dtype_limits` clipping behavior.
      Refactor code that generates scikit-image logo.
      Merge pull request #4 from stefanv/tonysyu_logo_refactor
      Merge pull request #468 from ahojnnes/rotate-doc
      Merge pull request #470 from ahojnnes/regionprops-example
      Merge pull request #477 from JDWarner/convert_dtype_to_astype
      Fix kernel-size calculation for non-zero theta.
      Add `bandwidth` parameter to `gabor_kernel`
      DOC: Tweak docstrings for Gabor filter
      Update gabor example.
      Add better test of gabor filter
      Merge pull request #371 from ahojnnes/gabor
      Merge pull request #498 from JDWarner/fix_canny_strel
      Merge pull request #497 from ahojnnes/guidelines
      Merge pull request #484 from almarklein/3d-selem
      Fix imread from urls on some backends.
      Merge pull request #512 from mkcor/doc_fix-deprecationwarning
      Merge pull request #509 from stefanv/contributor_guidelines
      Merge pull request #524 from stefanv/chelsea
      Minor cleanups
      Merge branch 'pr/517'
      Fixes to make doctests pass.
      Merge branch 'pr/531'
      Merge pull request #528 from chintak/doc_morphology
      Merge pull request #537 from ahojnnes/examples-fix
      Merge pull request #539 from ahojnnes/bresenham-deprecate
      Add `image_label2rgb`
      Fix: Add setup.py which adds data file to the build
      ENH: Add some checks on the inputs.
      TST: Add tests for image_label2rgb
      DOC: Change more examples to use `image_label2rgb`
      API Change: `image_label2rgb` changed to `label2rgb`.
      Merge pull request #536 from ahojnnes/edge-weights
      Add skimage.viewer.qt wrapper.
      Add PySide compatibility for save dialogs
      Fix file open dialog for PySide
      Remove `alignment` kwarg for compatibility with older PyQt4
      Merge pull request #541 from emmanuelle/doc
      Remove `alignment` kwarg (left out from previous commit).
      Fix `RequiredAttrs` definition.
      Fix parameter name in docstring
      Add infrastructure for conneting plugin output to a viewer.
      Add example of connected viewers/plugins
      Dock plugins to image viewer.
      Fix sizing of PlotPlugin
      Display title in docked plugins
      Add smart window-layout for multi-viewer display
      Add mock pyqtSignal to try to get Travis to build
      Second attempt to get Travis to pass
      Third attempt at getting Travis to build
      Accidental deletion
      Add attributes to Mock object to fix Travis build
      Fix PySide compatibility for signal defs
      Fix plugin interaction with CollectionViewer
      Add example of connecting plugins to CollectionViewer
      Fix display of overlay plugin when original image is updated
      Fix docstring for CollectionViewer slider.
      Change color definitions from ini file to python module
      Tweak Sphinx rendering of contribution page
      Merge pull request #593 from ahojnnes/rebase-workflow
      Merge pull request #590 from sciunto/python3fix
      Fix execution in IPython with qt backend.
      Fix PyQt4 segfault caused by PySide fix.
      Make histogram threshold adjustable
      Change QApp default to previous behavior.
      Raise ImageViewer to front on start.
      Change default Slider update_on value to 'release'
      Add explanation to LBP example
      Clean up test.
      Merge pull request #613 from adamw523/master
      Minor fixes
      Merge pull request #620 from scikit-image/2013
      Change rank.pyx to _rank.py
      Add support for all data types.
      Fix build script
      Merge pull request #615 from JDWarner/fix_lineprofile
      Merge pull request #618 from mkcor/filter_canny_passingtype
      Fix naming bug
      Fix Python 3 incompatibility (also, some wording tweaks)
      Merge pull request #657 from chintak/doc_fix_cmorph
      Merge pull request #667 from ahojnnes/denoise-fixes
      Merge pull request #726 from jni/fix-relabel-doc
      Add support for consistent color labels for sparse labels.
      Merge pull request #6 from stefanv/tonysyu-consistent-color-labels
      Plot cleanup and tweak to total PRs line.
      Add hough_circles change to API doc
      Label dates every 3 months instead of every month
      Fix whitespace
      Cleanups to imports and line wrapping
      Clean up novice __init__
      Change test class to test functions
      Speed up test
      Cleanup docstrings.
      Remove color scraping script and intermediate data file
      Change inflate/inflation to rescale/scale.
      Clean up docstrings.
      Remove unnecessary functions.
      Fix `modified` flag behavior
      Clean up color_dict definition
      Reuse colors from skimage.color.
      Rename novice package as per convention
      Switch argument order for clarity
      Clean up testing
      Fix validation bug and add test
      Rename image to array
      Remove color and size from Picture.__init__
      Remove `from_path` and `from_array` methods
      Simplify data access and add tests
      Remove unnecessary try-except.
      Simplify private docstrings
      Refactor and add error tests
      Rename for clarity
      Simplify indexing
      Change PixelGroup to inherit from Picture
      Move methods from PixelGroup to Picture parent class
      Move key validation for clarity
      Make xy_array public
      Remove PixelGroup dependence on pic and key attrs
      Fix setter for xy_array and reuse.
      Rename for clarity
      Remove PixelGroup and just return a Picture
      Fix doctests and cleanup `repr`s
      Replace use of elephant pic with Chelsea the cat
      Simplify interface
      Speed up doctest
      Remove old image reference
      Add docstring for arrays
      Rename image for clarity.
      Simplify docstrings
      Ensure that image shows when not run interactively
      Fix display in IPython notebook
      Remove extra newline
      Fix Python 3 incompatibility
      Fix the same Py3 incompatibility in a different place
      Fix Py3 range incompatibilities
      Add support for negative indexes/slices.
      Add support for non-unit step
      Fix ColorHistogram to work with Open File
      Merge pull request #802 from jni/allow-rgb-save
      Merge pull request #810 from jni/viewer-returns-overlays
      Fix hough_circle regression on windows
      Merge pull request #838 from ahojnnes/gallery
      Move image stack to its own module
      Move __init__ functions to submodules.
      Use explicit imports
      Remove debugging statement
      Remove unnecessary class
      Remove unnecessary initialization
      Refactor io doc building code
      Factor out table formatting code.
      Minor rename
      Fix wrapping of plugin descriptions
      Factor out config-file parsing for clarity
      Add comment for clarification
      Factor out url handling for unified file/url behavior
      Add test that error gets raised when no plugin available
      PEP8
      Fix test so it doesn't have side-effects
      Change InheritedConfig so a key is contained if a parent key is contained
      Move tests to separate file.
      Change plugin_store to use InheritedConfig
      Use InheritedConfig for plugin_store
      Add test for loading preferred plugins.
      Make preferred_plugins a dict
      Refactor load_preferred_plugins
      Move io._plugins.plugin to io.manage_plugins
      Modest attempt at normalizing file extensions
      Remove inherited config and file-type-specific code.
      Fix failing doctest on systems w/o PIL
      Slight clean up of tests
      Revert previous 3 commits to test Travis CI failures.
      Un-revert "Fix failing doctest on systems w/o PIL" commit
      Un-revert commits added to test Travis CI failures
      Fix doctest which causes side-effects.
      Test Image._repr_png_
      Test null plugin
      Attempt to fix failing test on Python 3 buildbot
      Add dynamic creation of imread_collection function
      Add comments about how plugins are managed.
      Move wrapper function
      Factor out function injection
      Remove outdated function
      Clean up test
      Merge pull request #881 from blink1073/fix-labelpainter-609
      Merge pull request #891 from sciunto/tuto2
      Merge pull request #895 from blink1073/fix-canny-py3k
      Add string inputs for range parameters of `rescale_intensity`
      Merge pull request #896 from ahojnnes/shared-six
      Add tests for uint12 and uint14 limits
      Merge pull request #959 from ahojnnes/viewer-backend
      Add example of tinting a grayscale image
      Incorporate PR feedback
      Fix a couple of grammar typos
      Change EPD link to Enthought Canopy
      Tweak range definition in `rescale_intensity`
      Make `intensity_range` a public function
      Edit wording of docstrings
      Clarify docstrings
      Clarify docstrings
      Add TODO for deprecation removal
      Add tests for intensity_range and rename parameter
      Use normal deprecation warning
      Add tests for intensity_range and rename parameter
      Merge pull request #893 from blink1073/fix-adapthist-757
      Merge pull request #1071 from blink1073/cleanup-adapthist
      Add `adapt_rgb` decorator and helpers.
      Remove cast and rescale and adjust test values to match
      Fix docstring for wrapped functions
      Remove orphaned import
      Merge pull request #1023 from ahojnnes/release
      Clarify release procedure doc
      Merge pull request #1065 from blink1073/multi_linetool_drawing
      Fix deprecation of canny function
      Add example for `adapt_rgb`
      Set colormap to grayscale
      Update import to match rename of `filters` module
      Fix compatibility with PIL
      Fix numpy 1.9 warnings about array.fromstring

Umesh (3):
      Updated Implementation of Roberts Algorithm
      Upadated Version with test added(/filter/tests/test_edge.py
      updated tests for Roberts Edge Detector

Victor Escorcia (1):
      Fix output type of peak_local_max

Vighnesh Birodkar (272):
      added blob_dog
      doctest
      added test_blob
      doc string
      typo
      threshold
      Changed formula, this improves detected size accuracy
      Made get_local_maxima_3d public
      changed params and functionality of get_local_maxima
      added get_local_maxima to feature
      removed get_local_maxima, now using peak_local_max
      rectified __all__
      added names in CONTRIBUTIRS.txt
      added blob_log
      added test_blob_log
      comments
      migrated from area to sigma
      removed print
      doc changes
      doc rectify
      typo
      overrided __eq__ method of region props
      removed blankline
      added tests, accounted for nan case
      added nan case
      pep8
      reverted irrelevant changes:
      reverted irrelevant changes
      reverted irrelevant changes
      reverted irrelevant changes
      removed array cast, added not equals test
      formatting, used assert_equals in test
      added cython file
      made methods private
      doc strings
      pep8 cython file
      added python call and doc strings
      removed print
      tests and minor corrections
      notes change
      bento change
      minor doc changes
      added full paper desc.
      corrections to cython file, switched dtype to double
      doc test
      indexing
      made hessian_det_appx public
      moved public interface of hessian determinant to corner.py
      doc string change
      extra line and spelling mistake
      improved test coverage
      changed parameter of hessian_matrix_det
      removed whitespaces
      removed blanklines in doc strings
      Added Hubble Deep Field image.
      Added example and pep8 changes.
      Corrected spellings.
      Grammar corrections.
      Grammar corrections.
      Changed travis.yml to install pillow.
      pip command correction.
      Tab to spaces.
      hubble_deep_filed doc string changes.
      Changed example to make code compact.
      Changed example and hubble_deep_field doc string
      autopep8 changes.
      Corrected comment in example.
      Added method reference to example.
      Added mean color RAG construction code
      Added edge threshold cut
      Renamed graph.py to rag.py
      Added testcase
      Added docs
      Docs formatting
      Added examples in docstring
      Formatting changes
      Minor Foramtting
      Corrected test mistake
      removed comments
      Added example
      Added explanation to example
      Added bento.info
      Added networkx to requirements
      Added networkx installation
      CHanges to get Python3 working
      Renamed _construct.pyx to _build_rag.pyx and some code corrections
      Removed cython file and added nd python function for RAG construction
      Corrected rag import
      reverted benfo.info and setup.py
      Added references
      type casting to double instead of long
      rebased and used color.label2rgb
      Added comments and switched try with if
      Added RAG merge_nodes test case
      added merge_nodes test
      Formatting changes
      Graph can now merge non adjacent nodes
      Used mapping in graph_cut.py
      Naming changes and formatting
      Added connectivity parameter and a function input to merge_nodes
      Made test Python3 compatible
      Formatting and typos
      removed commented stataments
      Formatting and docstring changes
      Switch label_image to labels
      Changed prototype of weight_func
      docstring corrections
      update test for py3
      None comparison
      Added RAG merge example
      Added comment about footprint
      Change merge_nodes code and test case
      Docstring changes.
      merge_nodes docstring change
      test_rag.py formatting
      Got Py3 test to pass
      Changed max_edge in example/plot_rag.py
      Use sets instead of dictionaries in merge_nodes
      Rename methods and edit docstrings
      Changes to plot_rag.py
      Typos
      Got doctest to pass and docstring changes
      docstring change
      Doctest Passing
      made RAG public and hyperlinked networkx.Graph documentation
      Docstring change
      Spelling correction.
      Fixed threshold in example
      First working copy
      pep8 changes
      Cleaned up rag.py
      cleaned up _ncut.py
      Docstrings
      Docstring of graph_cut.py
      rectified threshold mistake
      docstrings to _ncut.py
      docstring to _ncut_cy.pys
      Comments to graph_cut.py
      Added example for ncut
      Updated bento file
      used standard eignen solver
      doc string correction
      Remove unused except caluses
      use map_array instead of modifying graph
      rename method to cut_normalized
      Refer sections of paper in comments
      Add unit test
      Cut cost is computed by Cython code
      API changes, comments and docstrings
      rag copy in threshold cut
      test skip
      comment
      Fixed potential import error
      skip error test
      Docstring changes and typos
      docstring changes and code movement
      test case for in place
      docstring of _label_all
      corrected similarity
      Minor changes
      Fix Reference format
      change ref format in example
      Update _ncut_cy.pyx
      Update graph_cut.py
      Frist working copy of RAG draw
      Formatting
      Docstrings
      Corrected docstrings
      string wrap
      rebase and change API to support mpl colorspec
      None comparison
      Fixed doctest
      handled matplotlib not present case
      add desaturate option
      pep8
      Chnaged palette in example
      Chnaged palette in example
      removed file save
      corrected title
      fixed init
      fixed imports
      update relase doc
      force copy
      cast labels in ncut
      Added example with cubehelix
      remove self loops in rag
      Spelling error
      added non in-place merge
      Corrected new id logic
      change in random number max logic
      cleanup of if else block
      sequential numbering of nodes using max id
      Change None to 0
      always return new id and put new id logic in a function
      Docstring changes
      fixed blob area overlap function warning
      working code
      docstrings
      unit tests
      Description of example
      fix example
      rebase and docstring changes
      added nx check in test
      edge heap
      complete heap implementation
      blank line
      lists in the heap to avoid copying of dict
      doc string
      addition inline
      Rebase work
      Spelling
      spelling check
      Removed ASCII art
      handled the function with callbacks
      added merge_in_place option
      code cleanup and removed test
      for iterator change
      rename mapping array
      rename mapping array
      put code into functions
      changed function calls
      Added tests
      minor formatting
      Removed local debug statement.
      Improved test case and removed duplicate test
      Improved test case
      Fixed tests
      Added implementation and doc string
      Added example and test
      Formatting changes
      corrected title in example
      Added bento.info entry
      Corrected bento entry
      Removed debug print and corrected formatting
      Fixing error in ncut when subgraph as all equal weights (#1538)
      removed debug statements
      skip loop when min and max are equal
      Corrected Typo
      docstring change
      pep8 changes
      2d images converted to 3d before removing seams
      Corrected testing code
      Modifed code to use seam map
      added removal function
      Changed signatures to use energy map
      docstring updates
      Added rocket image
      Formatting changes
      pep8 changes
      Modfied test to use new API and added code to remove remaining seams if any
      Added code comments
      removed debug print in test
      Precomputation optimizations and docstring changes
      Changes example explanation text
      preallocated seam_buffer and added noil
      Changed return type to bint and improved code comments
      added boundary rag construction
      removed unnecessary statement
      indentation correction
      corrected colors import
      changed logic to use erosion/dilation
      passing selem argument
      changed to eosion/dilation logic
      cleanup
      doc example
      use as strided
      moved example file
      changed colormap to jet
      change mcap to viridis
      Merge pull request #1 from jni/edge-rag-fixup
      correct variable name mistake
      removed failing doctest line

Vincent Albufera (4):
      No interpolation when displaying small patches in example
      Addition of examples in the harris and peak local max function
      Addition of examples in the harris and peak local max function
      Addition of examples in the harris and peak local max function

Warren Weckesser (3):
      DOC: fix probabilistic_hough docstring--use correct keyword arg in the docstring, plus a few spacing tweaks.
      BUG: io: imread in the PIL plugin drops indexed PNG alpha channel.
      BLD: Increase the minimum pillow version to 2.1.0.

Xavier Moles Lopez (8):
      FIX: Ran autopep8 on colorconv.py FIX: modified color.data.ihc function FIX: smaller ihc.jpg image
      FIX: modified color.data.ihc docstring.
      FIX: import rgb2hed and hed2rgb in the test_colorconv script.
      FIX: Fixed test_hed_rgb_roundtrip.
      FIX: Link to Ruifrok reference
      ENH: Updated stain separation function docstrings and added matrices to __all__
      FIX: Example in combine_stains docstring.
      FIX: Added myself in CONTRIBUTORS.txt

Zach Pincus (29):
      Fix memory leak due to not properly using __cinit__ in heap classes
      Fixed (?) segfaults on 64-bit platforms
      use 64-bit safe types
      Fix another 64-bit corner case
      fix _spath error
      Add contour-finding
      Make it work
      Move to measure submodule
      add _find_contours.c
      Tighten up _find_contours.pyx
      Clean up void-pointer handling to clarify intent and fix multipage 64-bit bugs
      Make library-loading more robust and provide better errors
      Fix indentation error
      ENH: Add metadata reading functions
      DOC: Fix read_multipage docstring
      ENH: Clarify FreeImage IO flags, improve tests, and fix documentation
      ENH: load libfreeimage based on case-insensitive libname prefixes
      BUG: Make it work actually...
      STY: Clarify logic of code for FreeImage loading.
      ENH: Add tifffile IO plugin
      Fix description of tifffile plugin
      ENH: Add tifffile tests and improve freeimage tests
      Fix copy/paste error
      BUG: Fix tests
      BUG: MCP could segfault in places where both positive and negative moves would go out of bounds.
      Indent 4
      ENH: Add tests for boundary-overlapping offsets
      Fix 64-bit multipage tiff writing segfault
      Add multipage TIFF unittest

Zachary Pincus (12):
      Modify _build.py:cython() to be cleaner:
      Add fast binary heap classes (from Almar Klein)
      Added classes for minimum-cost-pathfinding through n-d arrays.
      Rewrite spath.shortest_path to use new mcp classes
      Finish commits of changes to graph: __init__ and setup and remove old cruft.
      Fix heap.pyx to work when infinite values are pushed.
      Performance fix for heap; correctness for _mcp
      Oops: actually fix heap performance problem...
      palette support for saving 8-bit greyscale images
      fix infinite-handling in mcp and precision issues in heap
      allow negative-cost elements
      add tests

andy (9):
      Convenience functions to load example images.
      hog: uniform filter instead of integral image
      pep8
      use dtype.img_as_float in _convert for colorspaces.
      bug: color convertion. make tests arrays float.
      don't use np.empty_like with dtype argument to be numpy 1.5 compatible.
      have preferrence list of plugins instead of just trying "null" and "pil"
      null_plugin gets imread as promised in __ini__.py of plugins
      pep8

arokem (3):
      DOC: Small one. The input to this one needs to be 2D.
      TST: Test error handling in convex_hull, convex_hull_object.
      TST: Test both functions(!).

arve0 (25):
      fix building of doc in virtualenv. add short description in readme
      fix github logo icon path
      use feature.canny
      fix typo artist.remove -> artists.remove
      move CollectionViewer `update_image` to parent class
      move docstring from viewer.widgets.core to viewer.widgets
      add note about latex
      import from _label instead of _ccomp
      install_requires from requirements.txt
      install sphinx before running tests in travis
      add some more magic
      close conda install code string
      add button class
      add reset button to crop plugin
      pop_bilateral docstring: the->pixels
      cpu_count may raise NotImplementedError
      deps: cython not runtime, add dask, six >= 1.4
      clean up setup.py according to 9cfaf4a
      update appveyor requirements.txt
      install binaries -> then resolve deps
      trust http wheel house
      use travis python 2.7 and wheel house
      smarter requirements in test env
      update install instructions
      expect PIL open fp

bennlich (3):
      Always maximally constrain bounding_shape when creating ellipse
      Fix bug when shape parameter is passed in from color images
      Explicitly cast upper_left and lower_right to ints

blink1073 (45):
      Add shaded polygon to lineprofile output
      Refactor perp_lines behavior
      Add test for LineProfiler tool
      Fix aliasing and update docstring
      Update test now that aliasing has been removed
      Use division consistently between PY2 and PY3
      Update test to use floating point images and with updated values
      Update the paint tool shape with the viewer image changes
      Add a shape property that gracefully accomodates a new image size
      Add viewer to main docstring.
      Add py3k compatibility for core widget and overlayplugin.
      Use the more idiomatic list().
      Add random seeds to tests per #1044
      Move random seed prior to first random() call
      Set the unused portion of the image to 0
      Implement a mode enum
      Update docstring
      Change 'trim' to 'crop'
      Add test for adapthist modes
      Switch to HSV, refactor equalize_adapthist, fix #757, cleanup docstring
      Update adapthist tests, add test for alpha channel and rgb mode handling.
      Fix failing tests and turn module level testing back ony
      Start removing optional edge pixel handling
      Fix handling of border pixels and update tests
      Refactor to use np.pad, naming changes, update test.
      Remove unused deprecation import.
      Minor cleanup of adapthist module
      Rename module to be more explicit
      Add tests for all of viewer subpackage, with necessary file mods
      Fix failing tests
      Address travis failures
      Another travis fix
      Skip tests if matplotlib is not present
      Make skip condition cleaner
      Move away from random.random in favor of random.rand
      Fix failing doctests
      Update Travis build to use Anaconda.
      Move handling of blit behavior to axes object.
      Push handling of line tools to axes level for better coordination.
      Remove unused parameter from docstring
      Refactor blit manager and event manager to viewer
      Switch plugins to use new tool api
      Updates to get all tools tested and working
      Make other suggested style changes
      Fix holes in lab conversions and add tests

capitanbatata (10):
      Added support for different illuminants.
      Added new tests. The support for a wider range of illuminants is now tested.
      Added tests for the new functionality.
      Undone change in skimage/__init__.py.
      The format strings were updated to include numbers.
      Corrected code to conform to Python coding conventions (PEP 8).
      Added Stephan comments.
      Removed the print statements from the test code.
      Incorporated Stefan's comments.
      Added Johannes' comments.

cgilon (1):
      Fix link to numpy docs convention

cgohlke (86):
      Fix ValueError on win-amd64
      Fix `TypeError: must use keyword argument for key function` on Python 3
      Work around `AttributeError: 'module' object has no attribute 'BufferedIOBase'` on Python 2.7+, Windows
      Fix `AttributeError: 'function' object has no attribute 'func_name'` on Python 3
      Fix watershed failures on Python 3
      Fix Python 3 import error
      Fix Python 3 import error
      Fix Python 3 import error
      Edited scikits/image/morphology/watershed.py via GitHub
      Edited scikits/image/draw/draw.py via GitHub
      Edited scikits/image/morphology/__init__.py via GitHub
      Fix ImportError on Python 3
      Fix: use npz instead of npy format for matlab-output files
      Fix: use npz instead of npy format for matlab-output files
      Fix: don't run all scikits tests when running scikits.image.test()
      Fix print statements missed by 2to3
      Update URL in package description
      Remove print statements
      Remove print statements
      Fix ImportError on Python 3
      Fix ImportError on Python 3
      Fix ImportError on Python 3
      Fix docstring use_plugin example
      Workaround 16 bit image to numpy array conversion bug in PIL
      Fix reading 16 bit big-endian images with PIL
      Rewrite dtype._convert function
      Merged some code paths
      Restore numpy 1.5 compatibility
      Restore previous io plugins
      Fix invalid Python 3 syntax
      Fix invalid Python 3 syntax
      Fix conversion of bytes to str on Python 3
      Fix TypeError on 64 bit platforms
      Add new 64-bit metadata types to METADATA_DATATYPE (new in FreeImage 3.15.3)
      Rewrite convert function
      Adjust tests for new type conversion rules
      Adjust rescale_intensity function to new dtype conversion rules
      Fix grammar
      Docstring additions
      Fix spelling
      Update docstring
      Fix whitespace
      Fix uniform conversion to signed int
      Fix rescale_intensity
      Fix numpy 1.7dev casting
      Verify dtype
      Return correct dtype
      Code cleanup
      Restore test_float_out_of_range
      Add floating point range check
      Fix TypeError
      Use general format
      Remove controversial range check
      Remove range check test
      Add range check against better judgment
      Add test_float_out_of_range
      Fix test failures on Python 3
      Fix build error on EPD for Windows
      Fix io.imsave() problems with passing arguments to plugin "freeimage"
      Open image files in default binary write mode
      Fix ValueError: Buffer dtype mismatch on 64 bit platforms
      Fix: round function is not provided by libc.math on all platforms
      Fix: round function is not provided by libc.math on all platforms
      Fix: 'inline' not permitted on data declarations
      Fix: 'inline' not permitted on data declarations
      Document correct parameter types
      Use io.BytesIO instead of StringIO
      BUG: basestring not defined on PY3
      TST: use BytesIO to save images into file-like object
      BUG: Fix ValueError: Buffer dtype mismatch, expected 'long' but got 'long long'
      TST: Fix ValueError: Buffer dtype mismatch, expected 'intp_t' but got 'long'
      TST: Fix test_weighted_moments is defined twice
      TST: Fix AttributeError: '_RegionProperties' object has no attribute 'weighted_moments_central'
      TST: Fix RuntimeError: data type not supported on Python 2.6
      unwrap: Fix crash on Windows
      unwrap: Fix crash on Windows
      unwrap: Fix crash on Windows
      unwrap: Fix crash on Windows
      unwrap: Fix unicode
      Fix ValueError: Buffer dtype mismatch, expected 'Py_ssize_t' but got 'long' on win-amd64
      Define M_PI (not part of C89)
      Define M_PI (not part of C89)
      Fix test_novice.test_update_on_save failure on Windows
      Use os.pathsep to sepatate multiple ImageCollection load_pattern
      Skip viewer tests if no Qt API is available
      Fix TypeError: __class__ assignment: 'TestFlexibleMCP' object layout differs from 'UnboundSelf'

dan (21):
      ellipse documentation was corrected
      more formal documentation
      pep line width violation corrected
      plain python multiblock lbp implemented with test coverage
      imports now inside of the file
      Cython implementation of MB-LBP. Updated MB-LBP visualization without matplotlib.Examples to gallery were added. Tests are made more easily readable.
      MBLBP optimized by taking out constants definition out of function. Visualization function of MBLBP now supports colors and opacity. Examples are updated.
      public MBLBP function created for users to safely use it.
      MBLBP is cdef function now. Corrected the example repeated imports.
      Switched to using integrate Cython function from skimage._shared. All changes were made to make it fit.
      small correction in documentation.
      moved multiblocl_local_binary_pattern in python file in order for sphinx to be able to correctly creat documentation.
      small documentation correction
      used gray2rgb function. Code is more readable now.
      pep8. Documenatation correction.
      Documenatation correction
      the function name was shortened to multiblock_lbp
      changed documentation
      Changed x/y to r/c notation. Documentation correction.
      Documentation correction.
      added missing visualization of the plot

emmanuelle (117):
      Initial integration of CellProfiler's watershed module
      Removed unnneeded code to compute strides
      Added the is_local_maximum function, that is useful to determine
      Import is_local_maximum in __init__.py
      New example for the gallery: watershed
      ENH: watershed docstring
      ENH: give a short definition of rank order in docstring
      Merge pull request #62 from tonysyu/fix_morph_symmetry
      Renaming: scikits.image -> skimage
      ENH: more efficient binary operation in skeletonize
      Merge branch 'neil_yager-skeletonize'
      First integration of cell profiler medial axis transform
      Medial axis transform (skeletonization) from cell profiler
      ENH address comments on PR #65 made by Stefan and Neil
      Moved cython file that had not been automaticall moved by rename
      scikits/image -> skimage in examples
      DOC: Canny filter. Improved docstring + example for the gallery
      Short tutorial on segmentation for the user guide +
      ENH contours of segmentation in segmentation tutorial
      BUG: greycomatrix, one of the tests was failing for symmetric case
      Merge pull request #1 from tonysyu/scipy_india
      Random walker algo for segmentation + example
      __init__.py in segmentation submodule
      More explanations about the algorithm
      Cross-See also in watershed and random walker
      ENH: addressed Tony's comments (renaming of functions, doc)
      ENH Removed unused argument in random_walker_segmentation
      ENH Tony's comments on random walker
      Merge branch 'random_walker'
      User guide: first draft of "getting help"
      getting_help file
      Merge pull request #162 from vincent-albufera/improve_doc
      Changed a link that pointed to a specific release of scikit-image using
      Implemented Stéfan's minor comments.
      Added a short paragraph + see also section in docstring of
      Fixed broken link to NumPy's docstring standard.
      First version of sections on color and exposure.
      Removed deprecated module name filter from tutorial
      Added an import of `median` from the `filters` module
      Added a second example to the edge filtering gallery example, focusing on
      Modified test for median filter import
      Added references on rotation invariance and filters.
      Completed docstrings for the different edge filters.
      Small additions to some docstrings in exposure module
      Corrected typos + modified docstring that made doctest fail
      Corrected another bug that made doctest fail
      Minor corrections
      Moved non-local means denoising to restoration submodule
      Modified bento file with non-local means extension
      Removed test file that has been moved to restoration submodule
      Modified import to be compatible with Python 3
      Faster version of non-local means denoising for 2D greyscale images
      Modified a few docstrings that made doctesting fail
      Enhancements to the text
      Modified the order of API docs page so that table of contents is always
      Over and underline alignment
      Removed seealso field that was not recognized in gallery example
      Updated numpydoc files from the github package numpy/numpydoc
      Minor corrections
      Minor corrections
      Removed normalization of h parameters because float images will be
      Implemented asymmetric distance computation to save speed factor of 2
      Renamed module, since tests were breaking because module and function
      Bug correction: I had forgotten to add __init__.py
      Implemented fast algorithm also for 3-D and 2D-RGB images. Changed API so
      Changed type of error returned for image dimension strictly greater than 3.
      Added docstring to helper functions in cython submodule
      Changed API of nl_means_denoising function to have a multichannel flag
      Merged 2D and 2D RGB functions for non-local means denoising
      Removed blank lines in docstring
      Changed variable names for coordinates for better consistency with
      [ENH] Removed eps parameter that was not needed.
      Global constant defined for distance cutoff
      [ENH] Docstring cleaning
      PEP 8: spaces and operators
      Helper functions for computing the integral of the difference between
      Modified variables' names.
      Modified name of image dtype.
      Removed a few redundant lines.
      Added comment for inner loops.
      Added a comment about normalization.
      Added some comments inside Cython functions.
      Added function to generate binary image with rounded (blob-like) objects.
      Removed obsolete Image class from io submodule.
      Addressed a few minor issues raised by Juan and Stefan.
      Used new function to generate binary blobs in existing gallery example
      Fixed example in doctest.
      Modified test with a less stringent condition on numerical value for the
      Modified (again) test accuracy.
      [ENH] Fixed some PEP8 issues in example gallery.
      Added test for slic segmentation: case where more segments than pixels
      Missing line
      Documented required version of sphinx
      Fixed broken sphinx link
      Some minor PEP8 issues
      Avoids crash of active contour example when run with old scipy version
      Added sections to gallery of examples
      Thickened contours so that they are more visible in gallery thumbnail.
      Moved active contours example in edges section
      Corrected bug in the computation of the Euler characteristic.
      Mentioned Euler characteristic (instead of Euler number) in docstring of
      Removed deprecated label function from morphology module in preparation
      Some PEP8 in test module
      Modified some methods in _RegionProperties class to make them
      Euler number is only2d
      Removed @property decorator, that made _install_properties_doc crash
      contribs.py: fixed the case of empty string
      Execute contribs.py in travis_script.sh
      Mentioned 2.6 hack for contribs.py in TODO.txt
      Minor fixes: errant indent, functools.wraps and python3 style
      First version of release notes for 0.12
      Minor change in order of contributors
      Correction in release notes (wrong section)
      3D test for register_translation, as required by TODO.txt
      Modified label function so that background pixels are labeled with 0, and
      Modified bento.info and skimage/__init__.py with 0.12.0
      Updated docversion.js

endolith (1):
      DOC: Change phase example from `jet` to `rainbow`

hawkerpl (1):
      fix more informative error while reading image

ivoflipse (4):
      Moved range_rows_start/stop and range_colums_start/stop to hog_histograms
      Fix an error in the centre calculation
      Fix an error in the centre calculation
      Reapplying the changes that got reverted

jaimefrio (1):
      BUG: wrong search for pivot in `quicker_sort`

kpk09 (5):
      Added broken test for feature.peak_local_max when the image is 3D and num_peaks is set. The current code only holds for 2D images.
      Fixed bug in feature.peak_local_max: num_peaks should now be usable for ND arrays.
      Using numpy.argwhere makes code easier to read
      Added Kevin Keraudren to the list of contributors: test and fix in feature.peak_local_max
      proposed fix and test for broken watershed on flat areas (https://github.com/scikit-image/scikit-image/issues/803)

kuantkid (4):
      FIX convert to (unsigned) integer from bool type
      FIX: early fix for dtype conversion from bool to other type PR #306, Issue #263
      FIX: multiply by the maximum value may change dtype
      FIX: issue sign loss warning only for signed type

martin (14):
      added variables to Makefile for build time configuration
      adding shared axes to examples with multiple plots
      changed adjustable parameter of axes to 'box-forced'
      adding axes sharing to displays of related images
      added sharex, sharey and adjustable parameters to subplots
      replaced subplots with individual subplot calls
      added parameters for axes sharing to plt.subplots
      added parameters for axes sharing to plt.subplots
      changed subplot creation to preferred matplotlib style
      changed subplot creation to preferred matplotlib style
      changed subplot creation to preferred matplotlib style
      changed subplot creation to preferred matplotlib style
      changed subplot creation to preferred matplotlib style
      reverted subplot creation to preferred matplotlib style

michaelpacer (3):
      fixed documentation to remove redundant links
      Fixed variable names to accord with r, c naming convention
      Added build location checker, does not fully work yet. Complicated error.

nich2o (4):
      corrected path.join in test_lpi_filter.py
      added a color image
      added rgb2hsv color conversion
      updated __init__.py in color

odebeir (30):
      initial commit
      paste code
      adjust code for new project
      fix pxd-pyx confusion
      restore setupfile
      fix pxd-pyx confusion
      restore setupfile
      compare bilateral
      rename rank.py to rank.pyx + adjust setup
      doc bilateral_rank
      doc
      start filter bigger example
      example continued
      doc cont.
      doc add perf.comp.
      document rank filters
      remove trivial examples from perentile_rank
      remove duplicate examples
      remove other duplicate examples
      local Otsu returns now the threshold values
      fix: Otsu(8bit) and bilateral functions docstrings
      fix:indentation
      add exemple to  denoise_bilateral() in docstring
      typo
      add gamma correction example
      add gamma correction example
      add entropy example
      pep8
      merge entropy examples+add description
      typos

oew1v07 (1):
      Merge pull request #2 from dpshelio/watershed_markes

pcampr (3):
      fixing multiple bugs in hog.py, adding two tests to test_hog.py
      fixed imports
      several fixes and 3 new tests for Histograms of Oriented Gradients

radioxoma (14):
      Add Yen threshold method
      Codestyle improvement
      Yen PEP8 and docstrings
      Added Yen tests, typo fixed
      Logarithm optimisation
      Fix import list sequence
      fix threshold_yen import
      fix yen blank image bug
      Add ISODATA threshold with tests
      contrib
      Unit-tests and small fixes
      fix tests
      Adding comments
      Explanation by ahojnnes

root (1):
      Updated function names[Robert's Edge Detection]

salvatore (8):
      Added an example of use of CanvasTools.RectTool to allow generation of docs.
      Added an example of use of CanvasTools.RectTool to allow generation of docs.
      Contributor file updated .... very nice to be part of this beatiful scikit!
      try to format properly a function in examples of RectTool
      deleted ">>>" in body of print_the_rect function
      ... added in indented code
      Travis auto-build Doctest skipped to avoid window opening
      removed numpy import and spaces

sccolbert (77):
      OpenCV infrastructure largely in place. A test function wrapped: cvSmooth()
      wrapped 7 more functions
      added unit-tests for existing functions. and prelimenary docstrings
      updated tests to be shorter and OO. Deleted unnecessary data files.
      added cvCornerSubPix and cvResize impementations. Added a bash build script.
      added a bash script for build to holdover until a setup.py solution
      added cvGoodFeaturesToTrack. Added clone_array_shape function.
      added a setup.py, moved the data files to the head folder.
      fixed the setup.py, it works now.
      Added cvFind and DrawChessboardCorners - from Holger
      added chessboard data images, modified install.txt, and .gitignore
      added 4 warping functions. Added CvMat support.
      added "cxcore" to __all__
      added camera calibration from Holger Rapp and fixed a bug in CvMat struct
      added morphology functions and support for IplConvKernel
      added cvFilter2D. This supports convolution with an arbitrary filter.
      verified only float32 is valid for cvFilter2D kernel
      synched with upstream and added integral images
      added color conversion constants
      fixed camera calibration to 4-vec for opencv 1 support
      bogus build commit
      added color conversion
      Merge branch 'cvtcolor'
      fixed camera calib back to a 5 vec since we support opencv >= 2.0
      added threshold and adpative threshold
      fixed threshold to add otsu as a kwarg
      added threshold and adpative threshold
      fixed threshold to add otsu as a kwarg
      added pyramid up and pyramid down
      added pyramid up and pyramid down
      standardized the handling of out arguments, replaced assertions with exceptions.
      1/3 of the doc-strings complete. Doc decorator created and added to a _utilies.py module.
      this is the actualy commit for _utilities.py
      changed how the doc decorator works. now it only adds the url.
      fixed an assertion that was preventing a slice of an array from being
      Added some more docs and fixed up a few warts in the process.
      more docs
      Finished the docs, and caught a couple more warts.
      Added the beginning of the gtk imshow plugin. And added the generic window manager.
      cleaned up the gtk plugin and window manager. rock and roll!
      cleaned up the window manager, and modified the qt-plugin to use it.
      removed the wx_plugin until which point wx supports pyos_inputhook
      removed the block= kwarg in the qt_plugin that i missed previously
      beginning of qt fancy imshow
      fixed event flow order for fancy imshow
      added the beginnings of the color mixes, and improved fancy imshow for qt
      Modified the fancy imshow widget, and added brightness control to the mixer.
      fixed a bug in slider signal connection.
      added slider value labels and fixed their layout a little.
      fixed an alignment centering bug.
      added HSV operations to the ColorMixer and fancy imshow.
      Finalized HSV support in the color-mixer and fixed a few subtle bugs therein.
      added histograms to fancy imshow and colormixer.
      added sigmoidal gamma
      progress commit
      further progress on the refactoring of the fancy imshow.
      added regular gamma to the mixer, refactored more of fancy imshow.
      Finished the refactor of the mixer and the histogram.
      fixed the gamma correction in the color mixer.
      updated string formatting for slider labels.
      Made it so that aspect ratio is maintained on imshow and fancy imshow for
      fixed mouse tracking to not go out of bounds. Added save-to-file.
      added save_to_variable functionality.
      Added save to variable function using io.push() functionality.
      removed a dangler.
      Added tests for colormixer and histograms. Moved tests up a directory level because
      first go at multithreading the mixer.
      added better logic to the thread dispatcher.
      added try except to deal with missing multiprocessing.
      added division decorators to cython. Improved how data gets chunked up.
      Fixed a small but serious bug in cvCvtColor.
      finished the lut tables
      added cvUndistort2 and its tests
      worked on prettying up scivi.
      added cvFindExtrinsicCameraParams2 and cvWatershed
      added cvFloodFill and cvMatchTemplate
      Merge pull request #18 from GaelVaroquaux/master

scottsievert (7):
      uses fftconvolve instead of convolve2d
      adds decision when to use fftconvolve/convolve2d
      fixes import, tuple addition
      adds comment explaining time_ratio decision
      now N dimensional, changes constant, cleans and comments
      removes paren
      styles code (no more functions, less lines)

sg (6):
      add bento to build docs
      added revisions by @stefanv and @ahojnnes
      updated gaussian doc
      added testing dependencies and standardized use of bullets
      remove extra :'s
      added ahojnnes revisions

sharky93 (1):
      Merge pull request #1 from stefanv/ipython-notebookcreation

spotter (17):
      ENH: Adding color deconvolution method to convert to an Haematoxylin-Eosin-DAB colorspace
      ENH: Added an Immunohistochemical image example
      FIX: the rgb2hed function. Legal issues regarding the ihc image. ENH: added the hed2rgb function.
      ENH: Added rgb2hed and hed2rgb to __all__ list
      FIX: Simplified deconvolution equations
      ENH: Added a short test to validate rgb2hed and hed2rgb conversion
      FIX: Removed unused dependency
      FIX: Testing for ubyte and float images. Input dtypes are converted beforehand.
      FIX: img_as_ubyte function threw ValueError due to floating point errors in hed2rgb function.
      ENH: Added example to the gallery for ihc staining separation
      FIX: Replaced non-ascii double quotes in Ruifrok reference.
      FIX: rgb2hed and hed2rgb can work with both signed and unsigned image.
      FIX: fixed docstrings. ENH: Added new stain separation matrices.
      ENH: Adding tests for new stain separation functions.
      FIX: using the new stain separation functions for the special case of hed.
      FIX: combine_stains docstring.
      FIX: changed unicode dash in docstrings.

stevendbrown (3):
      Added label vs. intensity shape checking to regionprops
      Changed to scikit-image preferred syntax
      fixed error language

sumitbinnani (3):
      Refactoring ```skimage.filters.gabor_filter``` to ```skimage.filters.gabor```
      Adding Alias and Deprecation Warning
      Adapting Edited Files to comply with PEP8 Standards

tangofoxtrotmike (1):
      feature.hog(): use y,x coords consistently, now works on non-square images

tv (2):
      The edge filter example contains all implemented filters apart from the Prewitt filter. Added the Prewitt filter to the example and compared it to the Scharr filter showing that Prewitt is even less rotationally invarient than the Sobel filter. Also added description.
      Show only Scharr filter result, removed plots of Sobel and Prewitt filter results. Reworked corresponding paragraph.

umesh (2):
      Final Roberts Edge Detector
      Updated Test Case

wilsaj (17):
      remove message about not being able to load nose; fixes #218
      raise an import error if trying to run test suite without nose
      add stefan's image class
      add html repr method for Image class
      add support for serializing to file-like objects (e.g. StringIO) to PIL plugin
      fix test so it looks for new Image class
      use repr_png and repr_jpeg hooks rather than repr_html
      allow qt_plugin imsave() function to write to file-like objects
      close temp StringIO buffers when we're done with them
      open buffer on on byte_array
      remove superfluous conditional logic
      remove unused import
      use standard conventions for Image.__new__() cls attribute and docstring
      role duplicate logic for Image._repr_png_() and Image._repr_jpeg_() into a common method
      add note about to display protocol to Image docstring
      add unit test for Image._repr_png_() with PIL plugin
      Remove tag metadata-related fucntions from Image class

yangzetian (4):
      Fix peak_local_max's output for ndarray when indices is set to False
      change doc for indices and remove the unnecessarily tolist() operation
      Merge pull request #1 from jni/peak-local-max-indices-nd
      Add test for nD `exclude_border` option

zachrahan (5):
      Merge pull request #1 from stefanv/find-contours
      Merge pull request #117 from zachrahan/freeimage-library_loader
      Merge pull request #149 from zachrahan/tifffile-plugin
      Merge pull request #507 from zachrahan/freeimage-multipage-64
      Merge pull request #508 from zachrahan/freeimage-multipage-64

Ömer Özak (1):
      Use np.ravel instead of np.copy

-----------------------------------------------------------------------

No new revisions were added by this update.

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/skimage.git



More information about the debian-science-commits mailing list