Channel Recombination Question

Although I’ve only worked with scripts so far, I know you can separate an RGB file into independent channels. When recombining them, can the process be used to align data to eliminate chromatic aberration from photographic lenses and lesser scopes?

Also, would there be an advantage to apply per-sub gradient removal on independent channels, or would that just mess up color?


Chromatic data is when one channel is a bit out of focus because of the lens quality (path length is different following the wavelength). So there’re no really solutions for that. You can try to reduce star size of the channel in another software and recombine after.

Siril already applies a gradient removal on independent channels.

Thanks for your reply. The reason I asked about the realignment is because I saw this video where there is a script for PixInsight that does exactly that, it will distort 2 channels to align them to a third, say align blue and red to green. So I was wondering if a similar process could be applied in Siril. It cerainly would be handy!

Thanks again for your reply.

If it’s a question of realigning the channels of course you can do it manually with siril, by extracting the channels, renaming them (or using convert) to make the files seen as a sequence, and using registration and RGB composition. But if the focus changes between channels, stars will always look bad.

Well then, I’ll assume Siril can’t distort the channels in the same way as the script in PI does so the point light sources are aligned, so that process can’t be carried out in Siril.


Vidrazor, I think you’re calling chromatic aberration something that isn’t. In chromatic aberration one or more channels are not focused in the same plane. This makes a point on the plane of focus a circular area instead of a point. This is not possible to fix by aligning the 3 channels, perhaps in some cases it could partially hide it (if you are lucky and the circular areas are hidden).

The pixinsight video describes the same technique that you’ve been told here before.

To remove chromatic aberration, you need to process with channels masked to affect only the aberrant channels.