Guided filter #1104

SamHames · 2014-08-17T22:02:24Z

This implements the guided filtering routine for edge preserving smoothing. The website is here: http://research.microsoft.com/en-us/um/people/kahe/eccv10/ . This is not based on the code provided there.

This is just the basic functionality with an example and (some) tests. I'd appreciate some comments and thoughts before I move any further with this.

Todo

Implement a transform.integral based averaging filter so runtime is approximately independent of window size.
Implement the option to use the covariances between colour channels as described in the ECCV paper.
Profile memory usage and give a better guide to the user.
Create some benchmarks and give a comparison with other options in skimage, such as the bilateral filter.

Some examples from the example script:

RONNCC · 2014-08-17T22:07:19Z

just making sure.. it's not patented/pending right?

On Sun, Aug 17, 2014 at 6:02 PM, SamHames notifications@github.com wrote:

This implements the guided filtering routine for edge preserving
smoothing. The website is here:
http://research.microsoft.com/en-us/um/people/kahe/eccv10/ . This is not
based on the code provided there.

This is just the basic functionality with an example and (some) tests. I'd
appreciate some comments and thoughts before I move any further with this.

Todo

Implement a transform.integral based averaging filter so runtime is
approximately independent of window size.

Implement the option to use the covariances between colour channels as
described in the ECCV paper.

Profile memory usage and give a better guide to the user.

Create some benchmarks and give a comparison with other options in
skimage, such as the bilateral filter.

Some examples from the example script:
[image: noisy_square_self_guide]
https://cloud.githubusercontent.com/assets/8400977/3945968/927b1906-2658-11e4-989c-04f22befb22f.png
[image: denoised_immunihistochemistry]

https://cloud.githubusercontent.com/assets/8400977/3945970/99b6916e-2658-11e4-8f44-7baa616f3a0b.png

You can merge this Pull Request by running

git pull https://github.com/SamHames/scikit-image guided_filter

Or view, comment on, or merge it at:

#1104
Commit Summary

Implement guided filter

Implement guided filter

Merge branch 'guided_filter' of github.com:SamHames/scikit-image
into guided_filter

Update examples and doc strings

File Changes

A doc/examples/plot_guided.py
https://github.com/scikit-image/scikit-image/pull/1104/files#diff-0
(120)

M skimage/restoration/init.py
https://github.com/scikit-image/scikit-image/pull/1104/files#diff-1
(6)

A skimage/restoration/_guided_filter.py
https://github.com/scikit-image/scikit-image/pull/1104/files#diff-2
(173)

A skimage/restoration/tests/test_guided.py
https://github.com/scikit-image/scikit-image/pull/1104/files#diff-3
(85)

Patch Links:

https://github.com/scikit-image/scikit-image/pull/1104.patch

https://github.com/scikit-image/scikit-image/pull/1104.diff

—
Reply to this email directly or view it on GitHub
#1104.

SamHames · 2014-08-17T22:19:33Z

Actually it might be. I'm having trouble wading through the patents - does anybody have any guidance on how to better do that?

coveralls · 2014-08-17T22:32:13Z

Coverage increased (+0.03%) when pulling b35adfb on SamHames:guided_filter into 9ea0d27 on scikit-image:master.

stefanv · 2014-12-01T12:53:36Z

Normally, the authors will clearly tell you if their work is patented. In this case, that doesn't seem to be the case. It is also included in the latest MATLAB: http://www.mathworks.com/help/images/ref/imguidedfilter.html

stefanv · 2014-12-01T13:00:31Z

@SamHames I browsed the paper and read your PR. First off, thank you very much for taking the time to contribute such a neat PR. Now, from the paper I understand that one of the benefits of this method is that it is fast, linear time--but this implementation is not, thereby losing that advantage. How difficult, do you think, would it be to change this to linear time?

SamHames · 2014-12-10T04:43:40Z

@stefanv Thanks for taking the time to have a look.

The only thing needed to make it constant run time is a box filter based on integral images/summed area tables. This is relatively straightforward, and can build on transform.integral_image . It would be best to have this implemented as a filter in it's own right as filters.fast_uniform (or something to that effect).

stefanv · 2014-12-10T10:44:22Z

@SamHames If you're up for it, let's do it!

sciunto · 2016-12-26T14:06:56Z

@SamHames I know that the last discussion happened two years ago. What's the status of this PR for you?

aizvorski · 2017-04-20T18:29:49Z

I would like to see this merged. Having fast box filters as the base for this is nice to have but should not be a show-stopper.

soupault · 2017-04-20T18:35:39Z

Let's do this.
@SamHames kind ping. Could you enable an access to the branch for the project maintainers (checkbox on the right of this page)?

SamHames · 2017-04-21T06:23:09Z

@soupault Done

Obviously life has intervened in the preceding couple of years :) Thanks for checking in.

I'll review where this is at sometime early next week - going from memory I don't think there's any significant work remaining, just details.

Implement the guided filter for fast, adaptive smoothing of images using local pixel statistics. Guided filtering is fast without approximations and computes the filtered image in one pass.

Implements the guided filter for fast, adaptive smoothing of images.

More informative examples for guided filtering now included.

pep8speaks · 2017-04-21T18:59:26Z

Hello @SamHames! Thanks for updating the PR.

In the file skimage/restoration/__init__.py, following are the PEP8 issues :

Line 32:80: E501 line too long (91 > 79 characters)
Line 48:1: E124 closing bracket does not match visual indentation

Comment last updated on April 23, 2017 at 13:50 Hours UTC

sciunto · 2017-04-21T19:27:57Z

doc/examples/plot_guided.py

+References
+==========
+
+.. [1] http://research.microsoft.com/en-us/um/people/kahe/eccv10/


This link looks broken... like redirected.

If I'm correct, the correct paper is: DOI:10.1007/978-3-642-15549-9_1

sciunto · 2017-04-21T19:28:42Z

skimage/restoration/__init__.py

@@ -11,7 +11,7 @@
       http://www.opticsinfobase.org/josaa/abstract.cfm?URI=josaa-27-7-1593

 .. [2] Richardson, William Hadley, "Bayesian-Based Iterative Method of
-       Image Restoration". JOSA 62 (1): 55–59. doi:10.1364/JOSA.62.000055, 1972
+       Image Restoration". JOSA 62 (1): 55-59. doi:10.1364/JOSA.62.000055, 1972


It would be better to write DOI instead of doi. and the year before the DOI.

sciunto · 2017-04-21T19:29:51Z

skimage/restoration/_guided_filter.py

+    References
+    ----------
+    .. [1] "Guided Image Filtering", by He et al. appearing in ECCV 2010.
+    http://research.microsoft.com/en-us/um/people/kahe/eccv10/


If I'm correct, the correct paper is: DOI:10.1007/978-3-642-15549-9_1

sciunto · 2017-04-21T19:31:15Z

skimage/restoration/_guided_filter.py

+
+def _guided_filter(image, guide, eta, radius):
+    """ Guided filtering on two distinct single channel float images."""
+    window_size = 2*radius + 1


Some PEP8 fixes to bring in these function.

sciunto · 2017-04-21T19:33:36Z

skimage/restoration/_guided_filter.py

+            [ 0.10548081,  0.10548081,  0.10548081],
+            [ 0.82811797,  0.82811797,  0.82811797]]])
+    """
+    if guide is not None:


Minor. as None is default, I would flip the position of this block with the elseblock to start with if guide is None

sciunto · 2017-04-23T10:59:22Z

@soupault What do you wish to do with this PR? Would you want to work on further enhancements or do you prefer that we consider merging this code?

soupault · 2017-04-23T11:10:13Z

@sciunto No, I just wanted to rebase the code, and perform the neccessary changes (PEP8-ify, refactor tests, move gallery example to the right place, etc). Pretty much done from my side (still need to fix failing docstring test).
I'd expect @SamHames to double check the implementation in the near time. Then we should consider merging the code.

codecov-io · 2017-04-23T22:53:15Z

Codecov Report

Merging #1104 into master will increase coverage by 0.04%.
The diff coverage is 98%.

@@            Coverage Diff             @@
##           master    #1104      +/-   ##
==========================================
+ Coverage   85.34%   85.38%   +0.04%     
==========================================
  Files         331      333       +2     
  Lines       26444    26543      +99     
==========================================
+ Hits        22568    22665      +97     
- Misses       3876     3878       +2

Impacted Files	Coverage Δ
skimage/restoration/tests/test_guided.py	`100% <100%> (ø)`
skimage/restoration/__init__.py	`100% <100%> (ø)`	⬆️
skimage/restoration/_guided_filter.py	`95.83% <95.83%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ad9db13...ea83790. Read the comment docs.

stefanv · 2022-10-20T07:07:37Z

Looks like we dropped the ball here. I am sorry about that, @SamHames.

stefanv · 2022-10-20T07:08:29Z

@soupault @grlee77 Any interest in salvaging this work?

soupault · 2022-10-22T13:53:32Z

I would say this is a classic algorithm and, definitely, is a good addition. I may pick this up later (no spare time currently). In any case, better to keep the PR open.

soupault added the new feature label Feb 6, 2016

soupault self-assigned this Jul 20, 2016

soupault added the 💪 Work in progress label Oct 25, 2016

soupault changed the title ~~WIP: Guided filter~~ Guided filter Oct 25, 2016

SamHames added 3 commits April 21, 2017 21:54

Implement guided filter

86d1937

Implement the guided filter for fast, adaptive smoothing of images using local pixel statistics. Guided filtering is fast without approximations and computes the filtered image in one pass.

Implement guided filter

aed8e4d

Implements the guided filter for fast, adaptive smoothing of images.

Update examples and doc strings

8eee365

More informative examples for guided filtering now included.

PEP8 for guided filter, switched to pytest

aa05f46

sciunto reviewed Apr 21, 2017

View reviewed changes

Fixed references, refactored code and example

dd6c601

Doctest: store output in variables

ea83790

Base automatically changed from master to main February 18, 2021 18:22

mkcor mentioned this pull request Feb 22, 2021

2021's calendar of community management #5169

Closed

grlee77 added 🙏 Feature request and removed type: new feature labels Feb 22, 2022

mkcor mentioned this pull request Feb 28, 2022

2022's calendar of community management #6165

Closed

SamHames closed this by deleting the head repository Oct 20, 2022

soupault added the 🆘 To take over label Oct 22, 2022

soupault reopened this Oct 22, 2022

soupault added 👶 type: New feature and removed 🙏 Feature request labels Oct 24, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Guided filter #1104

Guided filter #1104

SamHames commented Aug 17, 2014

RONNCC commented Aug 17, 2014

https://cloud.githubusercontent.com/assets/8400977/3945970/99b6916e-2658-11e4-8f44-7baa616f3a0b.png

SamHames commented Aug 17, 2014

coveralls commented Aug 17, 2014

stefanv commented Dec 1, 2014

stefanv commented Dec 1, 2014

SamHames commented Dec 10, 2014

stefanv commented Dec 10, 2014

sciunto commented Dec 26, 2016

aizvorski commented Apr 20, 2017

soupault commented Apr 20, 2017 •

edited

SamHames commented Apr 21, 2017

pep8speaks commented Apr 21, 2017 •

edited

sciunto Apr 21, 2017

sciunto Apr 21, 2017

sciunto Apr 21, 2017

sciunto Apr 21, 2017

sciunto Apr 21, 2017

sciunto Apr 21, 2017

sciunto commented Apr 23, 2017

soupault commented Apr 23, 2017

codecov-io commented Apr 23, 2017

stefanv commented Oct 20, 2022

stefanv commented Oct 20, 2022

soupault commented Oct 22, 2022 •

edited

Guided filter #1104

Are you sure you want to change the base?

Guided filter #1104

Conversation

SamHames commented Aug 17, 2014

RONNCC commented Aug 17, 2014

https://cloud.githubusercontent.com/assets/8400977/3945970/99b6916e-2658-11e4-8f44-7baa616f3a0b.png

SamHames commented Aug 17, 2014

coveralls commented Aug 17, 2014

stefanv commented Dec 1, 2014

stefanv commented Dec 1, 2014

SamHames commented Dec 10, 2014

stefanv commented Dec 10, 2014

sciunto commented Dec 26, 2016

aizvorski commented Apr 20, 2017

soupault commented Apr 20, 2017 • edited

SamHames commented Apr 21, 2017

pep8speaks commented Apr 21, 2017 • edited

Comment last updated on April 23, 2017 at 13:50 Hours UTC

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto Apr 21, 2017

Choose a reason for hiding this comment

sciunto commented Apr 23, 2017

soupault commented Apr 23, 2017

codecov-io commented Apr 23, 2017

Codecov Report

stefanv commented Oct 20, 2022

stefanv commented Oct 20, 2022

soupault commented Oct 22, 2022 • edited

soupault commented Apr 20, 2017 •

edited

pep8speaks commented Apr 21, 2017 •

edited

soupault commented Oct 22, 2022 •

edited