Master Flats - do you do noise reduction?

[mikey2000]

Here's another potentially daft question, once again illustrating my lack of depth-of-knowledge.

 

Background:  I've been persuaded to calibrate my ASI1600MM-Pro frames without Bias frames. Instead, I have a master Dark, and a master flat made from a large set of flat darks + flats.

As I understand it, the bias is unnecessary here as it is subtracted out of the flats already by the flat darks.   Also, the bias is removed from my lights by the master dark.  (I use exposure-length matched darks and lights)

So, I'm left wondering about the master flat.   I believe I'm left with a purely 'optical' master flat with no sensor bias or dark current to worry about.

 

Question:  Should I now do noise reduction on the flat?   

 

I believe the reason for shooting a large set of flats and DFs is to reduce noise in the master flat.   My optical path seems quite clean with only one large faint dust blob.   I can't help wondering if I'm wasting my time or not running a small scale NR to squash what's left of the fine grain in the master flat.

 

Just a thought....

[sloz1664]

In a word no. As you have removed the bias out of both the master dark, master flat, your calibration frames are complete. Just use your Masters to calibrate your images.

Steve

[vlaiv]

+1 for No.

CMOS sensors that have micro lenses and per pixel amps can have manufacturing defects - meaning not all pixels will have equal sensitivity to light. Flats take care of this, but if you denoise your flats you are running the risk of flats not correcting for those and thus having more noise in image with denoised flats than with pure ones.

[wimvb]

How many flat sub frames and flat dark frames go into making your master flats? Always aim for as large a number of sub frames for your calibration masters as possible.

With an ASI 1600 camera and short exposures, taking up to 50 sub frames for your masters, shouldn't take too long. The more sub frames that go into creating the calibration masters, the lower noise those will have.

But you shouldn't do any post processing of your calibration masters.

[mikey2000]

Thanks.  Point taken about cmos defects etc.  Perfect sense!
 

Wim - I used 50 each for making my masters (in fact, I think 120x for my master dark).  It is rather nice now owning a camera that I can plug into mains rather than constantly running out of batteries!