Is it possible to do a CMYK combine in Pixinsight

[goofisd]

I thought I saw a pixel math technique to combine CMYK monochrome images into a color image. But I can't find it now.
Does anyone know if this is possible?

Essentially, if you imagine I shot subs with a cyan, magenta and yellow filter, then integrated them into separate masters, how do I combine them to get accurate/semi-accurate color?

[oldwexi]

Very interesting.
Where did you get CMY astrophotography filters
Which camera delivers CMY Images

However,
My thinking to solve your question, maybe wrong,   is:
C consists of GB
M consists of RB
Y consists of RG

So logically the PixelMath Expression i would use is:
in the line of RGB/K:                  (Y*0.5) +  (M*0.5)
in the line of G:                         (Y*0.5) +  (C*0.5)
in the line of B:                         (M*0.5) + (C*0.5)

In the Destination area:
tag          create new image...
Set Color space to     RGB Color

Gerald

[goofisd]

Gerald, thanks for that - I was thinking along those lines as a fallback.

I never said I had the CMYK filters  ... I just suggested them for illustration.
(I'm working on a crazy narrowband issue and I think I can deal with it better in CMYK)

[Geoff]

Very interesting.
Where did you get CMY astrophotography filters

Just as an interesting aside I tried to see where these could be sourced.  Maybe here http://www.edmundoptics.com/products/sort_by_spec.cfm?productid=1734&sort=attrib_27563&order=DESC&accsort=accRoHS&accOrder=DESC
Geoff

[oldwexi]

@Geoff!
Thanks for the link to the filters.

@gofisd
Can you talk about the CYM narrowband Project?
I would be very interested to hear about these technics.

Gerald

[Geoff]

I think expressions of the following form are the way to do it
R=M+Y-C (=[R+B]+[R+G]-[G+B])
G=C+Y-M
B=C+M-Y

I have ignored the factors of 2.  I don't think they matter as long as you rescale your results.
Geoff

[goofisd]

Gerald, the basics are adding select Stromgren filters to my narrowband collection and getting better/more accurate colors in my narrowband imaging ... especially stars.

What I'd really like is being able to set the particular hue for each monochrome master and combine them based on what their center wavelength is. But I think that's asking too much (although someone has pointed me at some pixel math tutorials that just might make it possible).