Repository navigation
Docs: thin stacks underestimate Z drift, in Fiji Fast4DReg too - #8
Merged
Merged
Conversation
The docs said Z drift 'may not be found' on about 10 slices or fewer. Measured: it is underestimated, by a quarter to a half for beads and down to almost zero for smooth texture, and the original Fiji macro gives the same numbers on the same movies (0.02 / 0.74 / 0.99 for a true drift of 1 slice on smooth 10-slice, bead 10-slice and smooth 20-slice movies). The side view of a thin stack is a few rows tall; its fixed edges and Z profile pull the correlation peak towards zero. workflow.md explains it, with the table. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Scma6VXx7mT1FYsKmB1184
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The docs said Z drift "may not be found" on stacks of about 10 slices or fewer. Measured more carefully, it is underestimated, and the original Fiji macro gives exactly the same numbers. So this is a property of the method, not of the Python port.
Why. The side view of a 10-slice stack is 10 rows tall before stretching. Its top and bottom edges and its brightness profile along Z don't move with the drift, and a drift of 1–2 slices replaces 10–20% of the view. Both pull the correlation peak towards zero: tapering the edges with a Hann window moves the peak from 0 to 7 of the expected 12 px. NanoPyx alone finds shifts correctly on images of the same size.
Changes. README rule 3,
docs/workflow.md(the explanation and the table),docs/troubleshooting.md, andlc_testswording. No code change.Generated by Claude Code