GNU Astronomy Utilities - Bugs: bug #62702, Clumps on strong gradients of...
You are not allowed to post comments on this tracker with your current authentication level.
bug #62702: Clumps on strong gradients of larger galaxies become too large
Submitter: | Mohammad Akhlaghi <makhlaghi> | ||
Submitted: | Sun 03 Jul 2022 11:29:31 PM UTC | ||
Category: | Segment | Severity: | 3 - Normal |
Item Group: | Output not reasonable | Status: | In Progress |
Privacy: | Public | Assigned to: | makhlaghi |
Open/Closed: | Open |
Fri 04 Aug 2023 12:28:20 PM UTC, comment #4: |
Mohammad Akhlaghi <makhlaghi> |
Tue 01 Aug 2023 01:04:30 PM UTC, comment #3: I was able to start testing the hypothesis below by making the following figure: at the top you see the plot showing how the azimuthal angle changes as we come down from the brightest pixel of the small clump.
|
Mohammad Akhlaghi <makhlaghi> |
Mon 16 Jan 2023 11:29:38 AM UTC, comment #2: Until now, I had been investigating ways to step back to brighter pixels for finding a good stage to stop the growth of the clump. However, a few days ago, while investigating another issue with Sepideh Eskandarlou, we discovered a very interesting potential solution:
|
Mohammad Akhlaghi <makhlaghi> |
Sun 03 Jul 2022 11:35:44 PM UTC, comment #1: The image below is the output of the script in the original post:
|
Mohammad Akhlaghi <makhlaghi> |
Sun 03 Jul 2022 11:29:31 PM UTC, original submission:
When a clump falls on the gradient of a large object, in the current implementation of Segment, it will steal a large fraction of flux from the large object. The code below in the P.S can be put in a simple shell script to reproduce the problem.
|
Mohammad Akhlaghi <makhlaghi> |
Depends on the following items: None found
Items that depend on this one: None found
There are 0 votes so far. Votes easily highlight which items people would like to see resolved in priority, independently of the priority of the item set by tracker managers.
Follow 7 latest changes.
Date | Changed by | Updated Field | Previous Value | => | Replaced by |
---|---|---|---|---|---|
2023-08-04 | makhlaghi | Attached File | - | Added clumps-old-new.png, #55008 | |
2023-08-01 | makhlaghi | Carbon-Copy | - | Added -email is unavailable- | |
Carbon-Copy | - | Added -email is unavailable- | |||
2023-08-01 | makhlaghi | Attached File | - | Added azimuthal-angle-from-max.png, #55002 | |
2023-01-16 | makhlaghi | Attached File | - | Added azimuthal-angle.png, #54234 | |
2022-07-03 | makhlaghi | Attached File | - | Added large-clump.png, #53390 | |
Carbon-Copy | - | Added -email is unavailable- |
Powered by Savane 3.13-3230.
Corresponding source code
After some trials, in commit 5c96952b53d, I discovered that a sufficiently robust method is to use the variation of the azimuthal slice between subsequent pixels of a clump (as opposed to cycles).
In the image below, you can see an input image, and the resulting over-segmented clumps ('DET_CLUMPS_ALL' in the output of '--checksegmentation') between Gnuastro 0.20 and the commit above:
During the false clump identification the two elongated clumps on top of the good clump are rejected! Leaving us with a clean pixel map of only pixels that are strongly affected by that clump and not the wing of the galaxy.
This task is not yet finished and some extra tests in the next steps of segment (growth of true clumps in particular) are necessary.