Adding short exposure RGB's for star colour

[swag72]

I am looking at adding some short RGB exposures captured purely for star colour - Nothing else. I have looked at photoshop and this seems like a fairly simple process, to add a layer of the processed short subs for star colour (after all the images have been aligned), hide the layer and then just 'paint' in the stars that I want to add to my RGB image.

Is there a nice easy way to do this in PI? Would it be the HDR tool? Star colour is very important to me at this stage and I find that I want to increase the exposures of the RGB to get the detail, but of course saturate the stars in doing so. Hope there's a way to do this in PI.

[Josh Lake]

Instead of taking separate images to boost star color, why not make a Star Mask (with that function) and then apply it to the image? Then you can use ColorSaturation (or Curves) to intentionally bring the star colors out without affecting the background (with careful masking).

[vicent_peris]

Hi,

Are you sure you have a large amount of saturated stars? Or they are simply bright enough to not show the color? You can see in the linear images if they are in fact saturated. Anyway, I recommend you to try my technique explained in my article Dynamic range and local contrast. Read the third section, "High-Contrast, Small-Scale Structures". I think this is what you really need. After applying this technique, you can do what says JLake.

Regards,
Vicent.

[swag72]

Thanks all - Next time I am out I will measure the stars, but I think that at 600s RGB I have many fully saturated stars showing the max ADU. This is why I was thinking of a star layer of shorter exposures, then I could exposure for longer in the RGB as I'm not using Luminance.

If the stars are fully saturated, I will come back to the forum and ask again!!

[vicent_peris]

You can make then an HDR image with your short exposures. This is done with the HDRComposition tool. You'll need to smooth a bit the composition mask because the shorter exposures will have a different PSF than the longer ones.


Good luck,
V.

[swag72]

Thanks Vicent - What kind of smoothing would I be looking at?

I assume that the stars in the RGB image will be larger than in the star RGB image, so they will need to be reduced - Anything else?

[vicent_peris]

Thanks Vicent - What kind of smoothing would I be looking at?

I'm referring to the parameter Mask smoothness. It will depend very much on the PSF of both images; just experiment with it.

I assume that the stars in the RGB image will be larger than in the star RGB image, so they will need to be reduced - Anything else?

With a bit of luck you'll not need anything more... If so, please post some images to see which is the problem.


Regards,
V.