Repository navigation
Mask phase estimates without covariance support - #751
Open
scottstanie wants to merge 1 commit into
Open
scottstanie wants to merge 1 commit into
scottstanie wants to merge 1 commit into
Conversation
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change Summary
When a partially covered burst-edge cell has no valid samples for an acquisition, its coherence matrix has a zero diagonal. Phase linking can turn that empty estimate into unit magnitude and zero phase with temporal coherence 1, leaving seams in stitched interferograms. Mark these estimates as nodata while allowing a valid PS to fill them.
The fix uses covariance support rather than phase value, so supported zero-phase estimates remain valid.
On canis, a four-date, five-burst control using the unmodified development branch had 390 exact
1+0jpixels in each of six stitched interferograms. The upstream-based patch removed those pixels: 380 per interferogram were filled from overlapping bursts and 10 became nodata. Only those 390 pixels changed in each interferogram.Validation: 207 phase-linking, SHP, and PS tests passed on CPU (16 skipped); Ruff checks passed. On GPU, 206 passed and one existing CRLB tolerance test failed identically on unchanged upstream; that test passed on CPU.
Related issue number
#748
Checklist