RGBWS for creating and working with Luminance from RGB data

[lightwave]

Hi

I use a mono CCD, but I dont capture luminance separately as I have a fast system, so I capture each R,G,B at full resolution, bin 1x1.

My question arrises from the reading I have been doing to ensure I am using the correct RGBWS to be able to correctly extract a luminance and to use tools such as deconvolution correctly when I don't have seperate luminance data. I am about to process a large mosaic of the SMC and 47 Tucanae so I want to get it right, i.e. I dont want to do it again!

From reading other posts, below is what I believe is correct, have I got it right???

RGBWS has nothing to do with colour profiling, it is the space used for working with colour, calculating luminance etc.
Default is sRGB, which is very inappropriate for astro data, as far to much weight is given to the green channel. It was designed for terrestrial photography where the human eye is mostly sensitve to green and the objects are bathed in white light.

So I need to change the default settings to be able to work on and extract an appropriate LUMINANCE from RGB
Why is the default RGBWS set to sRGB?


From this

https://pixinsight.com/doc/legacy/LE/14_color_spaces/why_rgbws/why_rgbws.html


And this quote is from this thread by Juan from here http://pixinsight.com/forum/index.php?topic=4846.0

Deconvolution only makes sense for linear data. If applied to a nonlinear color component, such as a synthetic CIE L* component (lightness), then it is being used unrigorously with "cosmetic purposes", but not as deconvolution.

For this reason the Deconvolution tool offers you to work on the CIE Y component (luminance) or on separate RGB/K components: these are the *only* valid options for deconvolution.

However, special care must be taken to deconvolve the implicit luminance of a RGB color image:

- Of course, the raw integrated data must be used, i.e. the unstretched output data of the ImageIntegration tool.

- The RGB working space of the target image must be linear. Open the RGBWorkingSpace tool, extend the Gamma section, and uncheck the "Use sRGB..." option. Then set Gamma=1. Now you should define custom luminance coefficients. For compatibility purposes, the default weights are those of the sRGB space, which are quite inappropriate for DSOs (e.g., the green channel has ten times more relevance than blue and three times more than red). If you want to use a uniform linear RGB working space, then set the three coefficients equal to one. This is usually the best option. Now apply the RGBWorkingSpace process to the image.

- Select the Luminance (CIE Y) target in the Deconvolution tool.

- Use a suitable ScreenTransferFunction to inspect your image during deconvolution.



So I think what I need to do is the following, correct?

1-Change settings in RGBWS
   -Open RGBWS process
   -In Gamma section
         -Uncheck "use sRGB...”
         -Set Gamma=1
   -Set all Luminance (Y) coeffients =1.0, this gives uniform linear RGB working space
   -Leave the Chrominance settings
   -APPLY GLOBAL- these will be loaded on startup and affect all windows, ie I won't need to change them again

2-DECONVOULION and other sharpening tools
   -Select LUMINANCE (CIE Y) target in Deconvolution (and other) tool . (NOT LIGHTNESS)

Thanks heaps for any input.
Craig

[lightwave]

If someone is able to offer advice on this post i would greatly appreciate it, as Im going to process the mosaic this weekend

Thanks in advance

Craig

[pfile]

i don't think RGBWS is a global setting - you'll need to apply it to your RGB image before extracting the L*

but instead of doing that, why don't you make a synthetic L by integrating your 3 channel masters together? then you have a mono L image that's similar to what you would have gotten thru an L filter.

rob

[lightwave]

Hi Rob

Thanks for the advice!
Yes I had done that in the past, but after reading the posts listed above, I was thinking that the extracted luminance would give too much weight to the green channel as per the RGBWS. Does it use these settings for deciding what the “luminance” is, and if that is the case, would I notice any difference between the below 3 options?

-extract luminance from RGB master- apply deconvolution using dynamic PSF- combine LRGB
-apply deconvolution using dynamic PSF directly to the RGB master, target LUMINANCE (CIE Y)?
-applying deconvolution to each individual R G B master then combining.

Thanks again for your help (and anyone else!)

Craig

[lightwave]

Hi again Rob

I read your post again and realise what you mean is to integrate the masters as a Synthetic luminance. Good Plan!
I assume I need 1- No pixel rejection, 2-noise evaluation selected-to weight the best masters for the luminance.

Would you use any NB as well?, or just work on the RGB for Synth-L, adding NB later in the workflow

Thanks again
Craig

[pfile]

yes, when PI extracts L* it looks at the RGB working space settings attached to the image and weights the channels accordingly. so if you've set the gamma to 1 and the weights to 1, the extracted L* should not weight the green more than the other channels.

on the synthetic L, yes, actually the default settings for ImageIntegration are appropriate for making a synthetic L from RGB masters. by default it weights by SNR. i guess an argument can be made to include the NB in the synthetic L but i've never done that.

the deconvolution stuff is complex... if you are following an LRGB workflow where you process the L and RGB separately, there may be no reason to deconvolve the RGB. i dunno if you'd see a difference between those flows, you'd have to try them and see what happens.

rob

[lightwave]

Great!

Thanks heaps Rob
I really appreciate your advice
I'll let you know how I go with a couple of methods

All the best
Craig