PixelMath rescaling

[aworonow]

Hi,
  I just want to be perfectly clear about pixelmath's rescaling. I'm trying to track relative intensities across an emission nebula, and I want to account for camera-sensor wavelength dependencies and processing-introduced factors. One of those factors would be the rescaling of Ha vs OIII in pixelmath manipulations. The question is, when rescaling a RGB image, do all three colors use the same values for min and max? If so, are they the min and max of Luminosity, the K denominator (e.g. R/K, G, B) or are they different  for each band (Hopefullly not)?

Thanks, Alex

[vicent_peris]

Different.

[aworonow]

Actually, it appears to me that the combined largest dynamic range is used to rescale all three RGB channels. Here's what I did; let me know if I fouled it up somehow. Using pixelmath, I set R/K = random()+100, G = random(), B = random()-100. The output was set to be rescaled, 32 unsigned. The stat tool reported that the output image has...

                             R                          G                    B
minimum     9.950249e-001  4.975124e-001  0.000000e+000
maximum     1.000000e+000  5.024875e-001  4.975124e-003

Looks like the rescaling uses the ensemble (RGB) min and max...right?
 
Alex