Quote:
Originally Posted by Octane
I'll have a look at the JPM tekniq. The SHO palette was all green. I'm not sure what's going on. I linear fit everything to OIII as it had the broadest histogram. Perhaps I should have fit everything to hydrogen alpha? Any ideas as to what could be causing the massively green bias?
|
H,
The choice of LinearFit reference can make a difference but I haven't found a way to predict the best choice. I usually just pick one at random and work with the result.
A green bias is very common even after a LinearFit. DBE or BackgroundNeutralization will go a way towards fixing it. After a stretch I normally use the ColorMask script (not part of the standard distribution yet but you can get it on the PI forum or ask me for a copy) to do something equivalent to the Bob Franke method for PS (
http://bf-astro.com/hubblep.htm). Colour masking is actually more powerful than PS selective color because you can apply a range of tools to the masked image.
Removing the stars from the colour image (keep them in the synthetic luminance) works really well too. That's the core thing I took from JPM's tone mapping. It gets rid of those horrid magenta stars
It also allows you to do heavy noise reduction and a much bigger stretch on the weaker NB filters to get a lot more latitude with colour adjustment.
Cheers,
Rick.