DBE: correct only brightness, not colour

[TobiasLindemann]

Hi there,
is it possible to correct only the brightness in the background with DBE and let the colour gradients as they are?

Greetings
Tobias

[Juan Conejero]

Hi Tobias,

- Use ChannelExtraction to split your RGB image into CIE L*a*b* components.

- Apply DBE to the CIE L* component.

- Use ChannelCombination to regenerate your RGB image from the CIE L*a*b* components. Make sure to select the L* you have applied DBE to.

- Alternatively, you can use the HSI or HSV color spaces. These are actually what you should use if your image is linear, since CIE L*a*b* does not make any sense for linear images. In this case, I or V are the brightness components that you'd have to apply DBE to.

Just out of curiosity. Why are you trying to do this? (the actual reason to ask is: I am rewriting the DBE tool, so I'd like to be sure that I don't overlook something important).

[TobiasLindemann]

Hi Juan,

thank you very much for your detailed answer. It worked as it should :-), even if my problem was not solved.
My fault was, I thought DBE didn't work well on my image and it creates colour gradients. In fact the casts are already in the image and with DBE they just became more visible, because the background is more flat  .
Now I solved my problem by generating more samples (50 instead of 15 per row). In most cases fewer samples worked better for me, but in this case I had to create really much of it.
The third picture is with 50 samples per row and I am very lucky with it 
Do you have an idea, where these gradients (in first image) came from? Of course the pictures are wiiiiiide stretched to visualize the casts.

colour gradients


DBE with 15 spr


DBE with 50 spr


I like the DBE very much and I am very exciting for the new version  . As you asked for something that can be better: If you do a boosted STF while DBE, the samples became invisible. I think the boosted STF would be very helpful while selecting the samples.
normal STF:


boosted STF:


Greetings
Tobias

[georg.viehoever]

Hi Tobias,
I offen See similar color Patterns when Wirkung with my DSLR. They become more prominent when the Nights are warm, so i guess it is some thermal Noise Kasus.
Georg

[TobiasLindemann]

Hi Georg,

thanks, that sounds plausible. Although the temperature was not sooo warm. The camera-temperature was about 9 degree, so the outer temp. should has been about freezing point.

Greetings
Tobias