Cosmetic Correction Question

[Mark de Regt]

I think I have the basic idea, and I think I've created a decent cosmetic correction map (including bad columns).

But I still get some signs of bad columns in the integrated master lights.

Is this because I'm using master bias, dark and flats created without a cosmetic correction map?  If so, what's the fix?

Thanks!

[sreilly]

Your process should be to calibrate using your master frames and then use the Cosmetic Correction process. Then align and create the master light frames. Make sure you use the proper combine method depending on the number of frames you have. A guideline posted some time back was Percentile Clipping for up to 6 frames, Sigma Clipping for 7-10, Averaged Sigma Clipping for 10-14, Winsorized Sigma Clipping for +15, and Linear Fit Clipping for 25+.

I tend to use the default settings under Pixel Rejection (2) and always use Average for my Combination method. At least this is how I do it.

-Steve

[Mark de Regt]

I'm using batch pre-processing, and checking the box for using the cosmetic correction template I've made.  I assume that it applies it at the correct time.

[cfranks]

My camera is a QSI 540 and has hundreds (it seems) of bad columns but only 4 bright ones and those are the only ones I set in Cosmetic Correction.  My master Lights seem to be very clean after BPP does it's thing. 

Charles

[RickS]

I'm using batch pre-processing, and checking the box for using the cosmetic correction template I've made.  I assume that it applies it at the correct time.

Yes, BPP does run CC at the correct time.  Did you check the cosmetized subs to see how they look?  They should be in the output directory  (in subdirectory called lights/cosmetized or something similar.) Are you dithering?

Cheers,
Rick.

[Warhen]

Mark, I also like using a Super Bias created using the process of the same name. As Juan demonstrated a little while ago, these kinds of defects are better modeled there than in a normal bias, and will help with subtracting them as well as CC.

I think we should be careful not to call these weak discharges from single pixels, 'bad columns'- those are quite different.