Hi,
Despite looking for it for several days I fail to find a solution for it, so I assume it might be really technical and need to call for actual experts.
I acquire my sky images with a Unistellar Odyssey:
Buy a UNISTELLAR ODYSSEY smart telescope, today – UNISTELLAR EUROPE (include specs)
I retrieve the raw images as FITS, here is one for NGC 7023 (4s exposition, see FITS header for details):
20250508T022639_244_StackInput.fits (3.0 MB)
with the dark produced by the telescope during calibration:
20250424T201919_176_DarkframeMean (old).fits (3.0 MB)
(As new user, limited to 4 links, so I split here)
(continuation)
Here is my process:
- Convert all FITS images into a FITS sequence, around 5800 images of 4s
- Calibrate with provided dark (with debayer + calibrate all frames, bayer is GBRG in header)
- Background extraction on whole sequence
- Registration (global star alignment) after selecting a clear image as reference (reduce amount of registration failures, so more images retained)
- Stack (with weighted FWHM for weighting and same for rejection, keep 90%)
Then I have the stacked image (from 5058 retained images):
r_bkg_pp_c4_stacked-5058.fits (18.2 MB)
(renamed with FITS extension because Siril uses FIT which is not accepted here)
With an histogram auto-stretch on it + full color saturation (global, amount 1.00) it gets really red:
NGC 7023 is supposed to be a clear and nice blue, so I come back to original stacked image and go through:
- Astrometry (NGC7023, focal length 160.6, pixel size 1.45, should correspond to IMX415 sensor of Odyssey)
- Photometric color calibration with GAIA, because default
But it gives me something completely yellow when just displayed in autostretch:
With NOMAD catalog, it is neutral, but stretching and saturation like before gives an orange core:
Same thing with APASS catalog calibration.
It looks like I am missing something to get the proper blue color, but I fail to figure out what.
