SCNR question

Edward Lubat

Hi Adam

Is there a "right" time in the workflow to apply SCNR?  I am using a OSC camera (ZWO ASI071Pro) and not surprisingly have a greenish tint to my images that SCNR deals with very well.  But the reason I am asking is that I typically apply SCNR to my linear data right after PCC, assuming I now have the colors right and just have to "get the green out".  That usually works well.  But I recently was processing an image I took of the Rosette Nebula, and when I applied SCNR at that point, while doing the usual wonders to the background, it also turned the rosette from the proper red to an awful orange from which I could never recover.  I went back to just after PCC, skipped SCNR, and did the rest of my delinearization and postprocessing ignoring the green tint of the background.  Towards the end, while doing color saturation and manipulation of the stretched data, I applied SCNR which cleaned up the backround just as well but didn't seem to cause the same damage to the color of the nebula.

So is it better to apply SCNR before or after delineariztion?  Or does it depend on the data?

Thanks

Ed

BlockHead

Ed,

My gut reaction to your question is that SCNR shouldn't be used to take care of a significant green bias in an image- as you know it will change the color balance. So, I think you are doing it in the right order/place ... though I wonder if you should apply DBE first...then SCNR. But as you say, the green bits might be easier to deal with later in non-linear. Still... in an incremental processing I would expect you should be able to do SCNR while things are linear without any "damage."  I am not as familiar with OSC behavior (and the green pixels of the bayer matrix could be preferentially getting more weight because there are more of them). 

I am saying all of this blind without seeing the image... so, if there really is a bias- I think you need to attack this first before doing SCNR. Otherwise, it sounds like you are doing what is best.

-the Blockhead