StarTools

I've tried using the combine module but don't really understand how to set the weights, as I'm combining for example Ha to red and then using this as the new red channel.
My goal is to enhance the colors of the horsehead LRGB image with the addition of the HA and Sii data.
Here is what I have:
L = 37920 sec.
R = 13350 sec.
G = 8940 sec.
B = 7980 Sec.
Ha = 47220 Sec.
Sii = 25200 sec.

Thanks!

Hi,

You may be interested in this thread as well.

The short answer is that there is no one prescribed way to add narrow band to visual spectrum data. As such I cannot really give you any guidance on exposure time weightings of narrow band vs visual spectrum.

You can add narrow band as luminance, as color (red channel) or both.

For the purpose of colouring, you can add Ha, SII and visual red together in any way you see fit and make that the new R channel, after which you would process LRGB as usual.

In that thread I linked to, I also highlight a way that does not incorporate narrow band at during normal processing, but only adds it afterwards by processing the narrowband separately and layering it in using the Layer module.

Both flavors of techniques are popular. The latter is more advanced and flexible, but more involved and less fool-proof.

Ivo,
Thanks for the quick response. Using the techniques you mentioned I was able to add color with the layer model easily.
The image lacks detail though. I used the Develop module and set the same Digital Development percentage to compare the different data stacks. (is this a valid way to compare data?)
From this exercise, it's clear that the L channel lacks any real detail. The Sii channel is not as bright, but much cleaner and far more detailed. The Ha is also very clean and detailed but a bit dimmer still than the Sii. Now I don't know if that exercise really tells me anything other than the level of detail since they are quite different exposure lengths and filters. (see first post for exact exposure times)

Do you have a suggestion for adding detail? Adding one of the NB channels (saved as red in the compose module) by using the layer module added some extra detail along with the color, but I know there is more detail that can be squeezed out. I've tried the compose module loading L and Ha into R and G, set to L+syn L, mono. The result was not much different than the L by itself.

Thanks!

The key is to always process luminance (detail) and chroma (color) separately, so this will never be an issue.
If you like the colour you get from a particular combination of bands for your coloring, there is nothing stopping you from processing detail (luminance) separately with any other combination of bands you see fit.

You don't have to use the Compose module/mode either (though it should yield the cleanest results). You can also use the old school method of processing luminance and color in sequence, and then combine them in the Layer module (set Layer Mode to "Color of fg" or "Brightness of fg").

Does that help?