Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Use thresholded binarize and mask filtering in existing watershed code. #1671

Merged
merged 1 commit into from
Dec 6, 2019

Conversation

ttung
Copy link
Collaborator

@ttung ttung commented Nov 22, 2019

Refactor the existing watershed code to use the new binarizer and mask filtering (for binary opening).

Depends on #1670, #1651
Test plan: still see 69 cells in the iss notebook.

@codecov-io
Copy link

codecov-io commented Nov 22, 2019

Codecov Report

Merging #1671 into master will decrease coverage by <.01%.
The diff coverage is 86.36%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master    #1671      +/-   ##
==========================================
- Coverage   89.85%   89.84%   -0.01%     
==========================================
  Files         239      239              
  Lines        8962     8964       +2     
==========================================
+ Hits         8053     8054       +1     
- Misses        909      910       +1
Impacted Files Coverage Δ
starfish/core/image/Filter/util.py 81.81% <ø> (-1.52%) ⬇️
starfish/core/image/Segment/watershed.py 95.79% <86.36%> (-0.7%) ⬇️

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 339eaa9...85a77c0. Read the comment docs.

@ttung ttung force-pushed the tonytung-thresholding branch 2 times, most recently from 819db0d to a174408 Compare December 3, 2019 20:39
@ttung ttung changed the base branch from tonytung-thresholding-base to tonytung-binarize December 3, 2019 22:05
@ttung ttung changed the base branch from tonytung-binarize to master December 5, 2019 00:52
Refactor the existing watershed code to use the new binarizer and mask filtering (for binary opening).

Depends on #1670, #1651
Test plan: still see 69 cells in the iss notebook.
ttung pushed a commit that referenced this pull request Dec 5, 2019
Uses the labeling algorithms provided by #1680 and the area filter from #1673 to implement labeling.

Depends on #1671, #1673, #1680
Test plan: ISS notebook yields 96 cells.  The previous implementation did not support 3D and flattened everything along the Z axis.  Processing in 3D exposed issues in `peak_local_max`.  If we use the footprint + exclude borders approach, there is an off-by-one error in trimming the Z axis, resulting in completely blank images and no peaks.  Therefore, we have to exclude the borders.  Because of that, we detect more cells.
@ttung ttung merged commit 8440731 into master Dec 6, 2019
@ttung ttung deleted the tonytung-thresholding branch December 6, 2019 18:48
mattcai pushed a commit that referenced this pull request Dec 19, 2019
Uses the labeling algorithms provided by #1680 and the area filter from #1673 to implement labeling.

Depends on #1671, #1673, #1680
Test plan: ISS notebook yields 96 cells.  The previous implementation did not support 3D and flattened everything along the Z axis.  Processing in 3D exposed issues in `peak_local_max`.  If we use the footprint + exclude borders approach, there is an off-by-one error in trimming the Z axis, resulting in completely blank images and no peaks.  Therefore, we have to exclude the borders.  Because of that, we detect more cells.
ttung pushed a commit that referenced this pull request Jan 13, 2020
Uses the labeling algorithms provided by #1680 and the area filter from #1673 to implement labeling.

Depends on #1671, #1673, #1680
Test plan: ISS notebook yields 96 cells.  The previous implementation did not support 3D and flattened everything along the Z axis.  Processing in 3D exposed issues in `peak_local_max`.  If we use the footprint + exclude borders approach, there is an off-by-one error in trimming the Z axis, resulting in completely blank images and no peaks.  Therefore, we have to exclude the borders.  Because of that, we detect more cells.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants