Active Regions image is non-figurative
The active regions image looks non-figurative, with unexpected features
ver: 2.9.1 dev win11, 64 GB ram
in iron (for the Helium) it looks good
in Ha it is so-so
Found another one in the meanwhile
ok...
PTAL at https://melix.github.io/blog/2025/02-01-jsolex-actve-areas.html for a beginning of explanation and limitations
I think the internal reflections may be a contributing factor, it eerily resembles the flat's and reflections' position
I need to figure out some filtering mechanism, this will improve over time, I hope!
I think you could go back from the assembled disk:
take the final disk apply a large radius median, like 100 pixel (for the disk of 3000px diameter) subtract from the median the disk image, sunspots turn bright, everything else goes dark mark the areas considerably darker than the median (ie the highlights of the result of the subtraction) with the coordinates of the above, go back to the scanlines: if, at the expected scanline there is a vertical feature, that's a bingo for a sunspot and not a filament.
for active regions with no sunspots, the (diskImage - median) should be of interest