Background colour in single shot colour imaging

[ASTROMONK (David)]

Please can someone explain to me how to achieve a black background colour in a single shot colour image? I use Nebulosity camera control (SX M26C ) and can save the image as a single Maxim FITS frame or separate R G B FITS (Maxim).

[irwjager (Ivo)]

Originally Posted by ASTROMONK

Please can someone explain to me how to achieve a black background colour in a single shot colour image? I use Nebulosity camera control (SX M26C ) and can save the image as a single Maxim FITS frame or separate R G B FITS (Maxim).

Hi David,

Your question is quite broad - there are many things that can cause background levels. Some are easy to deal with (a constant level) while others are much harder to deal with (vignetting, light pollution and/or gradients). They all require different techniques to get rid of them.
Perhaps you could attach an example of the unwanted signal?

[ASTROMONK (David)]

Hello Ivo..

Thanks for reply....the background is a constant, even colour tint, usually mauve. I am familiar with light pollution and use a TLS (Astronomik) filter when this is a problem. M51 looks orange with or without the filter!
I took images of the Orion region with a Tak. Epsilon These were fine. I tried M51 with the 10" RC and came across this problem. The RC is much slower, f8 as opposed to f3.3 for the Epsilon. So the signal to noise is worse. It could be just that, as I am trying to extract signal from the noise floor??

[irwjager (Ivo)]

Originally Posted by ASTROMONK

Hello Ivo..

Thanks for reply....the background is a constant, even colour tint, usually mauve. I am familiar with light pollution and use a TLS (Astronomik) filter when this is a problem. M51 looks orange with or without the filter!
I took images of the Orion region with a Tak. Epsilon These were fine. I tried M51 with the 10" RC and came across this problem. The RC is much slower, f8 as opposed to f3.3 for the Epsilon. So the signal to noise is worse. It could be just that, as I am trying to extract signal from the noise floor??

Getting rid of an even bias during image processing shouldn't be too hard - just use a pedestal in the red and blue color channels using a curve manipulation tool of your choice.

It'd be better to eliminate the cause of it during acquisition though.
Of course, a scope itself should not induce any type of bias, so the culprit has to be elsewhere. Normally I can think of a few causes but these don't explain the different results you're getting with 2 different mirror based scopes. The only difference I can see is the Tak having a built-in field corrector lens (if I'm not mistaken?), which may interfere with how any (present or absent) IR/UV cut filter would influence the image. A slight magenta cast is consistent with an absent IR filter.

To be honest though, I can't see anything wrong with the image you attached - background levels are perfectly fine when I inspect their RGB values?

[ASTROMONK (David)]

Thank you for your help Ivo.
I found one cause of my problems yesterday, which may help other people also.
I take the image using Nebulosity. Then I attempt to open it up in AstroArt. It is at this point that the orange, or mauve cast is introduced.
So now I use only Nebulosity and the results are MUCH better.
There seems to be some difficulty - which I dont understand- about transferring colour FITS files.
Live and learn!
The attached file of M51 was processed in Nebulosity. Taken with the 10" RC, plus Intes focal reducer. I managed to clout the telescope halfway through the exposure! Background is a bit light, but its a balance between that and getting as much detail as possible from the subject.

[irwjager (Ivo)]

Originally Posted by ASTROMONK

Thank you for your help Ivo.
I found one cause of my problems yesterday, which may help other people also.
I take the image using Nebulosity. Then I attempt to open it up in AstroArt. It is at this point that the orange, or mauve cast is introduced.
So now I use only Nebulosity and the results are MUCH better.
There seems to be some difficulty - which I dont understand- about transferring colour FITS files.
Live and learn!
The attached file of M51 was processed in Nebulosity. Taken with the 10" RC, plus Intes focal reducer. I managed to clout the telescope halfway through the exposure! Background is a bit light, but its a balance between that and getting as much detail as possible from the subject.

Interesting... Maybe a bug in AstroArt? Glad you got it sorted though!
Nice image too. You should be able to get rid of that background glow though.
One easy way is to use PS or the GIMP, create a second layer and fill it with the orangy/background color (just use a dropper to sample the RGB value of a background pixel that seems like a good representative). Now subtract this layer from the original.
You've now subtracted the background level. The image is a bit dimmer now though, so use some curve manipulation to push up the levels a bit and voila!
Or you could use something like GradientXTerminator, PixInsight ABE/DBE or StarTools Wipe.

[ASTROMONK (David)]

Thanks again Ivo....please excuse my ignorance, but what is PS and GIMP?? Which software package do they appear in?

[irwjager (Ivo)]

Originally Posted by ASTROMONK

Thanks again Ivo....please excuse my ignorance, but what is PS and GIMP?? Which software package do they appear in?

Sorry, after a while you start talking in acronyms and abbreviations...
PS = Adobe PhotoShop, which is a great (but expensive) generic commercial image processing program.
GIMP = The GIMP (GNU Image Manipulation Program), which is a free open source equivalent of Adobe PhotoShop.

They both allow you to layer multiple images on top of eachother and/or perform other tricks like subtracting images (layers) from eachother.
Give the GIMP a try - it's free!

[ASTROMONK (David)]

I had some problems with a colour cast in the centre of the frame on my SX M26C camera. This has now been fully resolved by Terry Platt who sent me some new camera firmware! Great support - thanks Terry!

[grantboxer (Grant)]

I am using both Nebulosity (for image capture) and AstroArt (for processing) and I do not get any colour cast. I use a QHY-8 single shot colour camera - so I would guess it was your camera? It is important to check the bayer matrix style, as this can vary from camera to camera. If it is set incorrectly you can end up with some wild colours!
Cheers Grant

[multiweb (Marc)]

Unless you are imaging under extremely dark skies you will always get a color cast in your background, especially if you use filters such as LP. UHC-S (Red cast), Fringe Killer (yellow cast), etc... You just get rid of it at processing time. CCDStack has a 'desaturate background' option when you color combine. I assume other softwares would have the equivalent.