New Feature : Stellar Colour Calibration for Live Stacking

[admin]

Hi Roel,

I'm not seeing a big difference with those images, but I do see a noticeable change with the original image that you sent me (M 22_4.0s_356s_NEXQ.fits). See the video below (hope it plays) showing how the fringing is reduced around a lot of the stars when the option is enabled.

2025-08-26 14-53-48.mkv

(1.43 MiB) Downloaded 37 times

I think you need to be able to see the red/blue fringes on opposite sides of the stars for it to have much effect - that offset is what allows the code to measure how much to move the colour channels, and if the fringing is small/missing then the re-alignment will have practically no effect.

cheers,

Robin

[MarMax]

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

Robin, hopefully this link will work. It should be the three FITS files and text files.

https://www.dropbox.com/scl/fo/kystwllb ... tbdq4&dl=0

Mike

[admin]

Hi Mike,

thanks - those are interesting as they are wide-field images, which I wasn't expecting to have issues with...

It looks like the code that tries to automatically work out 'how close' the image star co-ordinates and the catalog star co-ordinates have to be to each other is choosing a stupid value (0.67 arc seconds) for those images, when it should probably be about 10 arc seconds. This means that most of the matches that should be found between image stars and catalog stars are being discarded Hopefully a relatively easy fix now I know what to look for!

cheers,

Robin

[roelb]

Thanks Robin for the latest version.

I did a lot of tests, with the old and new SC version, with the "Folder Monitor Camera", on previous taken images.
When the moon is disturbing the "observations", the blue color is still partly seen on the image after SCC.
But a much lower value with the latest version.
This seems to be expected because the moon light is disturbing unevenly the exposure image area.

But with no moon the results are very good.
After the SCC the histogram shows that the blue color is now aligned with the other two colors.
In some cases very small blue histogram corrections may be necessary but overall the SCC delivers good images.

I'm very satisfied with the improvements on the SCC.

Thanks again.
Roel

[MarMax]

Robin,

I've been consistently using the SCC tool and with the Edge11 kit I'm getting a lot of blue stars. Here are some examples.





Is this happening because I'm incorrectly adjusting the background?

[admin]

Hi,

the SCC should auto adjust the background for you, and even if you make tweaks to that then it should have little impact on the colours of the brighter stars as they are *so much* brighter than the background.

The best way to diagnose this sort of things is to first make sure that you have the very latest version of SharpCap 4.1 (as I did add some changes at one point to reduce the blueness). If the problem is still there with the latest, send me an image from live stacking 'Saved as 16 bit stack' - that lets me reproduce the issue. In fact you can also mess around using this image by loading it into the 'Test Camera 1 Deep Sky' and enabling live stacking - that image will be replayed every 8s and added to a stack (which will obviously never change as it is getting the same image over and over), but you can then do testing with SCC or other adjustments in the way I would when testing.

cheers,

Robin

[MarMax]

Hi Robin,

I've been spending time on a couple of different EAA kits and need to get back to the Edge 11 at f/7 to provide some 16-bit FITS stacks. I've been consistently using SCC and would like to know if your recommendations in Post #32 are still applicable with the latest tweaks to the code. Here is my take on the SCC best practices:

- allow 2-3 minutes of subs to stack before running SCC
- make R-G-B adjustments in Background Subtraction*
- run SCC again mid stack
- make R-G-B adjustments in Background Subtraction
- for final fine tuning make small adjustments to the live stack histogram R-G-B sliders

*the SCC sets the live stack histogram R-G-B sliders based on the appropriate star colors so changing these settings upsets the balance

It's a bit tricky to make fine adjustments to R-G-B in Background Subtraction. A very small adjustment goes a long way. Is there some keyboard or mouse command that will allow finer resolution movements? I use a Logitech ERGO M575 trackball mouse.

Thank you,
Mike

[admin]

Hi Mike,

yes, that would be my suggestion - the histogram colour balance adjustments are the ones that are scientifically calculated by the calibration and those values are pretty reliable. The background on the other hand has a much simpler estimation calculation that could be tricked by things like vignetting or large areas of nebulosity (filling most of the frame).

As with a lot of slider things in the live stacking area, you can make finer adjustments to the background if you start dragging one of the sliders and then (still holding the mouse button down) move the mouse vertically away from the slider so it is further up the screen. Now resume dragging left/right and you will see that the slider moves a smaller distance than you are moving the mouse, giving you finer control.

cheers,

Robin

[MarMax]

Robin,

I finally had a cloud free night to run the Edge 11 f/7 kit. Here is a link to the 16-bit FITS for M2.

https://www.dropbox.com/scl/fo/kystwllb ... l1esc&dl=0

Hopefully this will help with the extreme blue diagnosis. This is with the latest SC drop.

Mike

[admin]

Hi Mike,

thanks for sharing - I see two distinct things going on with the colour calibration in that image

1) the background has a red tinge in the center of the image and blue towards the edge/corners. I think that this is caused by uncorrected vignetting of the image, meaning that there isn't really a consistent background level - this in turn makes it hard to get the background subtraction just right when using the calibration. Cutting the image down to the central region (keeping 50% of width & height) cuts out enough of the vignetting on the edge/corners that the calibration then does a good job of getting a black background.

2) The calibration seems to be boosting the blue channel too much, giving a blue globular cluster. I need to dig into this one a bit more - is it the result of relatively noisy data in the B-G calibration graph? Or could it be due to the bulk of the calibration stars in that area being the brighter (and hence perhaps bluer) stars. If I can find some previously saved globular cluster images then I will test them out to see if I see similar things on other globs - if not then I can see if I can capture some on a clear night (if we get any soon...)

cheers,

Robin