[bug #62702] Clumps on strong gradients of larger galaxies become too large

[Mohammad Akhlaghi]

Follow-up Comment #1, bug #62702 (project gnuastro):

The image below is the output of the script in the original post:

(file #53390 large clump)

On the left, you see the image before adding noise. In the middle, the image
with noise is shown. On the right, you can see the pixels associated with the
two clumps as the white and yellow regions. The blue pixels are the undetected
sky regions, and the black pixels are the detected, but diffuse regions that
couldn't be associated with any individual clump.

As you see above, a small star has been placed very close to the larger
galaxy's center. 

We expect the "clump" of the star to only be as large as the black circle that
the two white arrows point to. However, because it has the gradient of the
background galaxy under it, Segment's current algorithm will associate the
full right-side of the larger galaxy to also be part of this clump.

We need to define a way to stop the growth of the smaller clump in such cases
and let the galaxy take the pixels that belong to it.


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?62702>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/