Calibration of Light Frames (CMOS)

D

dpaul

Well-known member
I'm using a ZWO ASI 2600 Pro and when the light frames are calibrated they appear to be much noisier (grainy) compared to the uncalibrated frames. This is how I'm doing the calibration based on advice seen on this forum and elsewhere:
(1) MASTERBIAS - this is an integration of 100 (0.03ms frames) with no normalisation. The calibration box is not checked when filling in the calibration settings. (2) MASTERDARK - the settings are exactly the same as the lightframes (temp, gain, binning) and the calibration box 'is' checked and the optimization level is set at 3.0. (3) I don't use flats.
During calibration of the light frames, there are no problems regarding correlation between master dark and target frames.

I have read previously where masterbias is best avoided for CMOS imaging so I tried calibrating the light frames again without the master bias. Now it keeps saying no correlation between master dark and target frames.

Can anyone help on advice on how best to calibrate CMOS frames without getting a grainy/noisy appearance compared to the uncalibrated frames?

Thanks

David
 
dpaul said:
I'm using a ZWO ASI 2600 Pro and when the light frames are calibrated they appear to be much noisier (grainy) compared to the uncalibrated frames. This is how I'm doing the calibration based on advice seen on this forum and elsewhere:
(1) MASTERBIAS - this is an integration of 100 (0.03ms frames) with no normalisation. The calibration box is not checked when filling in the calibration settings. (2) MASTERDARK - the settings are exactly the same as the lightframes (temp, gain, binning) and the calibration box 'is' checked and the optimization level is set at 3.0. (3) I don't use flats.
During calibration of the light frames, there are no problems regarding correlation between master dark and target frames.

I have read previously where masterbias is best avoided for CMOS imaging so I tried calibrating the light frames again without the master bias. Now it keeps saying no correlation between master dark and target frames.

Can anyone help on advice on how best to calibrate CMOS frames without getting a grainy/noisy appearance compared to the uncalibrated frames?

Thanks

David
Click to expand...
In addition to my previous note, I've now stopped checking the optimization level (so without using Master Bias it now doesn't have any correlation issues between Master Dark and the light frames. The result is still grainier than the original light frames. I noticed the Master Dark integrated frame looks grainy, maybe that's the issue?
 
See https://pixinsight.com/forum/index.php?threads/for-beginners-guide-to-pis-imagecalibration.11547/ , chapter 7.1, case 2.

This requires that the MasterDark is generated by simply integrating the dark frames. In your case (no flat frames) no MasterBias is needed.

Do you use fiters? Which exposure time are you using?

Please check whether the calibrated light frames are clipped in the low range (see https://pixinsight.com/forum/index....is-completely-left-on-all-images-stuck.17227/ , post #5).

Bernd
 
The general recommendation for CMOS cameras is to use flat-darks* for calibration.

What I do is to generate a master flat-dark and use this to calibrate the flats and generate a master flat. Then I use this master flat and a traditional master dark to calibrate the lights (Calibrate and Optimize options disabled during the whole process).

*A flat-dark is simply a dark that has the same exposure time as the flats.
 
bulrichl said:
See https://pixinsight.com/forum/index.php?threads/for-beginners-guide-to-pis-imagecalibration.11547/ , chapter 7.1, case 2.

This requires that the MasterDark is generated by simply integrating the dark frames. In your case (no flat frames) no MasterBias is needed.

Do you use fiters? Which exposure time are you using?

Please check whether the calibrated light frames are clipped in the low range (see https://pixinsight.com/forum/index....is-completely-left-on-all-images-stuck.17227/ , post #5).

Bernd
Click to expand...
Hi Bernd, thanks for the reply. Yes that's how I produce the masterdarks. This is a mono camera so I'm using CMOS optimised Baader LRGB filters. This is a 30'' scope so I only need about 8-12 sec frames (depending on binning) or up to 16 secs when using narrowband. I did actually try integrating the flats with zero pre-calibration and it didn't make much difference to the final image except a bit brighter (because no darks). So I will continue to calibrate but with zero bias frames.
Thanks
David
 
Marcelofig said:
The general recommendation for CMOS cameras is to use flat-darks* for calibration.

What I do is to generate a master flat-dark and use this to calibrate the flats and generate a master flat. Then I use this master flat and a traditional master dark to calibrate the lights (Calibrate and Optimize options disabled during the whole process).

*A flat-dark is simply a dark that has the same exposure time as the flats.
Click to expand...
Thanks for the reply - I don't use flats because I have to drag the 30'' scope out each night onto an equatorial platform. When returning to its cabin after imaging I can't leave the camera in position so can't do flats the following day. It has superb Lockwood optics and I get zero vignetting so no need for flats. See my reply to Bernd just now, I'll now continue with darks to calibrate the light frames but no bias master bias or optimisation any more.
Thanks
David
 
Hi David,

It is not expected that (correctly) calibrated light frames are noisier than the uncalibrated light frames. Maybe this is results from different Auto STF stretches?

If you are willing to upload a light frame and the MasterDark to a filehoster and post the link here, I will take a look at it.

Bernd
 
Hi Bernd
Apologies for not replying until now -
I'm happy with the results now without using bias frames - the ZWO ASI 2600 gives low noise frames.
David
 
You must log in or register to reply here.
Back
Top