I’m using a ASI183MC pro camera for capturing images. Capture software used is astroberry and the Images are saved as FITS. The FITS header state RGGB as Bayer pattern and Siril (0.99.8-1) is set to use the pattern in the header.
I would expect the images to be more green, because of the two green pixels compared to only one pixel for blue and red. But the image is more blue (single image, auto stretched). So I’m not sure if the color channels are swapped in a way and if I force any other Bayer pattern it comes out even stranger. If I set BGGR then the image is more red what is expected, because just red and blue are swapped out.
Any ideas on that? I can provide a sample image if needed.
Mario.
Most white balance and demosaicing systems automatically correct for this, which might be why it doesn’t appear unusually green.
Have you tried converting the FITS to TIFF using imagemagick (which does a straight conversion without white balance multipliers), and shooting something with “known” color properties (might be hard to shoot a terrestrial colorchecker with this setup???).
Wait a second… That alone wouldn’t quite work since that would convert to a greyscale TIFF. Sorry about that, I’m a bit rusty from some of my work with converting FITS to DNG and tagging to feed to RawTherapee…
I would shoot something of known color (e.g. solid blue object, solid green, solid red) to check for obvious channel swapping.
In general with recent versions of siril, the bayer pattern is correctly interpreted by Siril. But some capture software write FITS files upside-down and don’t document it, in that case the bayer pattern needs to be reversed with the two last characters first.
If you find this is the case, you can send this link to the authors of the capture software: Siril:FITS orientation - FreeAstro
Thank you and good luck.