New Feature : Stellar Colour Calibration for Live Stacking

[Gfischer]

Hi Robin,
I just uploaded the newest sc. I ran the scc using the hi resolution files. The color correction is perfect even with my lquad filter. You’ve done it again! Two minor notes.
First if you close the livestack while the scc is trying to plate solve it continues to solve and notification bar doesn’t have a cancel button.
Second the auto stretch is still very strong.
I really appreciate you getting scc to work so well with my quad band filter. Great job

[roelb]

Hi Roel,

glad that the fainter moonlight is helping - are you using any filtering for these images? Just wondering if one of these dual/tri/quad band filters might help for some of the targets (obviously mainly the nebulae). I had a go myself with a quad band filter a few days back and did the same target both with and without the quad band filter - the SCC gave similar final results in both cases, which I think is promising.

I had a look through the new images you shared, which do look better, but one of the Barnard148 images struck me as odd - a clip from it below

Screenshot 2025-08-14 110618_c.png

There are various dotted lines in primary colours here - you'd think they would be hot pixel trails, but that doesn't make sense as they are not all parallel to each other (even with field rotation, there would be a turning pattern in the trails, not the sort of random direction effect seen here).

I can't quite work out what is going on - you might think they would be cosmic ray trails or similar, but then why would they be dotted, and in one colour rather than affecting all colours. Weird...

cheers,

Robin

Hello Robin.
I searched the Barnard 148+149 image with the dotted colored lines.
It was one of the images re-produced with the Folder Monitor Camera, during daytime, using the fit files from the night session: https://mega.nz/file/tbNUmYAR#BvXiEdFkM ... GnOis39I60.

Another image, also made by the Folder Monitor Camera with the same fits files:
https://mega.nz/file/BSEXQbyT#ql4XFxNhC ... Sz6i4C47dU
doesn't have those trails.
I can't remember which setting was different between those 2 runs.

Also the "save with adjustments" image during the night session:
https://mega.nz/file/5SklzY7C#FJ6USJJp1 ... Lwy3lQs_PI
doesn't have those trails.

I inspected a lot of other recordings from the latest sessions and can't find such anomaly.
That includes sessions using the SC version 4.1.13584.0 and the latest version 4.1.13621.0.

Conclusion:
Such anomalies do not occur generally.
Their is no further investigation necessary

I hope this clarifies the issue.

Thanks for your appreciates efforts to make continuous improvements to the excellent SC software.

[roelb]

OK. I found the issue of those trails:
when I don't select Hot Pixel Removal Only, those trails show up.
So it seemed that they are indeed hot pixels.
Note: I don't use darks.

[Robert]

Hi Robin,

I used Test Camera 1 to experiment with the new Stellar Color Calibration. It works magnificently. Quite a solution!
I noticed that Hi-Res (199MB) found 23 star matches and succeeded. Ultra Hi-Res (622MB), however, found 13 star matches and failed.
Ultra Hi-Res has solved many of my plate solving challenges. Might you be adding the additional star color data also to Ultra Hi-Res?

Thanks, Robert

(RE: Barnard 148 images. I too had weird trails on my stacking images until I started using Hot and Cold Pixel Removal.)

[admin]

Hi folks,

currently the ultra high res doesn't have star colour data, but I could add it (it would just grow by about 25%). The latest versions of SharpCap will check the high res database for colour and use that if it is there (the 199Mb version), otherwise they will use the default one with colour. This is regardless of your plate solving index choice in the settings. I am wondering if the difference between 23 and 13 stars was more down to either star detection or plate solving co-ordinates being slightly different (which could change how many of the detected stars are close enough to the reference stars to be used).

If anyone has data that is consistently having problems with not enough stars with both the latest (4.1.13621) version of SharpCap *and* the 199Mb high res plate solving index with colour installed, please let me know (and share some data with me to test with!)

Going back to Roel's blue background - I have tweaked the way that the correlation curves are calculated to give more weight to the points near to the chosen 'reference white' colour - this pulls the curve closer to the data near the point where it is going to be used, and should help somewhat.

cheers,

Robin

[MarMax]

Robin,

What kind of data do you need? I was just at a dark site (SQM-L 21.6) using UV/IR cut filters only and was surprised that colour calibration failed several times. This was in massive star fields with lots of recognized stars. I only save PNG's and sometimes 16-bit FITS. Also, is there a way from the text file to identify if SCC was used? I'm asking because I don't remember the images where SCC failed.

Mike

[admin]

Hi Mike,

SCC gets recorded in the log file - if it works you will see lines like this

Code: Select all

Info    16:32:58.196078 #24 Using star colour data from C:\Program Files\SharpCap 4.1 (64 bit)\psindex.bin                                          in string SharpCap.ImageProcessing.PlateSolving.SharpSolvePlateSolveFactory.FindIndexFilePathWithColour()
Debug   16:32:58.379570 #24 Image centered on RA=00:42:43.7,Dec=+41:15:58, 927 detected stars have valid colour information                         in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Info    16:32:58.379824 #24 Loading plate solving index PSINDEX;VER=1.2.0.0;SOURCE=GAIADR3;CREATED=2025-06-20T22:20:54.6236750Z;MINGRIDSIZE=0.250;TILES=522;MAGDATA=1;COLORDATA=1, data size = 52.959Mb                 in new SharpCap.ImageProcessing.QuadAlignment.MultiTileBase(Func<Stream> streamSrc)
Debug   16:32:58.386685 #24 Catalog data provided 19843 nearby stars with colour data recorded                                                      in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Debug   16:32:58.388962 #24 413 reference stars with colour are inside the image field of view                                                      in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Debug   16:32:58.393817 #24 Found 344 stars with matching catalog entries. Positional match threshold 1.79 arcsec                                   in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Debug   16:32:58.414767 #24 G/R colour analysis on 200 data points                                                                                  in bool SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateGRMagOffset(List<ValueTuple<ICatalogEntry, StarInfo>> matches)
Debug   16:32:58.433924 #24 Best fit for G/R data is Grouped Linear L.S.D. with r^2 = 0.746, fit line is 0.00000 x^2 + 0.52003 x + -0.50581, r2=0.746, sse=0.002                    in bool SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateGRMagOffset(List<ValueTuple<ICatalogEntry, StarInfo>> matches)
Debug   16:32:58.434580 #24 B/G colour analysis on 200 data points                                                                                  in bool SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateBGMagOffset(List<ValueTuple<ICatalogEntry, StarInfo>> matches)
Debug   16:32:58.456435 #24 Best fit for B/G data is Grouped Quadratic LSD. with r^2 = 0.996, fit line is -0.63312 x^2 + 1.41216 x + 0.19955, r2=0.996, sse=0.013                   in bool SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateBGMagOffset(List<ValueTuple<ICatalogEntry, StarInfo>> matches)
Debug   16:32:58.457391 #24 White background image for G/R is -0.203 mag (reference is 0.583), multiplier 1.20518                                   in Nullable<(double red, double green, double blue)> SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateWhiteBalanceFromAnalysis(IColourReference refWhite)
Debug   16:32:58.457544 #24 White background image for B/G is 0.588 mag (reference is 0.321), multiplier 0.58204                                    in Nullable<(double red, double green, double blue)> SharpCap.ImageProcessing.PhotometricColourCalib.PhotometryColourCalibration.CalculateWhiteBalanceFromAnalysis(IColourReference refWhite)
Debug   16:32:58.457667 #24 Colour balance multipliers calculated : R=1.2052 G=1.0000 B=1.7181          

In particular, note the last line which gives the colour balance multipliers that have been calculated. A failure will instead end something like this in the log

Code: Select all

Debug   16:32:15.282807 #8  Found 0 stars with matching catalog entries. Positional match threshold 0.92 arcsec                                     in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Info    16:32:15.282874 #8  Not enough matches found to perform colour calibration                                                                  in CalibrationResults SharpCap.ImageProcessing.PhotometricColourCalib.FullPhotometryCalibration.Execute(ICapturedFrame frame, float alignCentile, IMappingData mappingData, IColourReference referenceWhite)
Error   16:32:15.282986 #8  Failed to get white balance factors from photometry colour calibration                                                  in async Task SharpCap.FrameTransforms.LiveStacking.ColourCalibrationViewModel.DoAutoAdjust()
Debug   16:32:15.284580 #1  Notification (Status=Error): Cannot perform colour calibration - No valid result was produced  

All I need to test this out is the stack saved as '16 bit' from the live stacking save button (top option of the four). That gives me basically the same data that would go to the SCC routine.

cheers,

Robin

[MarMax]

Robin,

I've found three 16-bit FITS stacks where the SCC failed. These are 150Mb files so how do I get them to you? And do you also want the text files that go with each?

Mike

[admin]

Hi Mike,

the best way with big files is to upload them to some sort of online file sharing tool (Google Drive, DropBox, things like that), then use the sharing functionality built into whatever platform you choose to share the files via a link which you can send to me (either post here, or use the 'contact us' link at the bottom of the page to send via email). The text files are handy to have too, since they are small anyway and might be useful. No hurry mind you, as I am away this week, so won't be able to do anything with the files until at least next week.

cheers,

Robin

[roelb]

Re. The new Enhancement option to re-align colour channels in deep sky Live Stacking.

I don't see any difference between selecting this new option.

Examples: https://mega.nz/folder/ADswhATR#xvM1nWMENWxlhzmd9tT-JA