Pixinsight - DBE - Before of after Channel Combination?

[Jbro1985]

I am going through the motions of properly processing my first narrowband image.  Perhaps there is no difference here to LRGB, but I am curious as to which method people favour for their work-flow and why:

A:  Combine the channels (using SHO palette in my sitution) and then apply DBE; or

B:  Apply DBE to each channel and then combine?

Mainly interested in nebulae.

Thanks in advance for any thoughts on this.

[RolandKol]

 just experiment... I am almost sure it will depend only on the quality of your images...

PI is really flexible in this case.

I usually use it before and after.

usual start: ABE on each channel, combine, ABE again, crop, - save the crop. I place DBE points manually on this combined image, - save icon. Discard the combined image.

Undo all actions on NB images if any, use saved Crop and use saved DBE on each (+deconvolution, sharpening, noise reduction and etc), once combined, use saved DBE again.

My location is really light polluted and this way works as a good start.

P.S. almost all tutorials advise to stretch all NB channels before combining them, but in my case, I always failed to calibrate star colors after.

but if I combine liner/unstretched ones, - colors become manageable... No I idea why...

[Jbro1985]

Yea - I played about with the image the other night just very quickly.  Didn't stretch before combination.  Combined, stretched via unlinked STF and Histo and then created colour masks to eliminate the green and then magenta and all seemed pretty good on the colour side.  I'm going to go with combining before stretch.  I also want to work on Noise Reduction whilst the image is still linear.