Dark subtraction problems...

[The_J]

Hi friends!

I´ve tried to make my first narrow band image processing. I have hera my Darks, Lights, Bias and Flatframes, made a MasterDark, MasterBias (not SuperBias) and MasterFlat. I calibrated all Lights and got a strange picture as result: It seems that the dark hasn´t subtracted from the image, the Amp-Glow (ASI183MM Pro) is strong visible! I tried several times to reject the possible error by variations of the parameters and resetting the files (making some sets of Masters). But I don´t get removed the amp glow???

I have attached examples of my trial picture (Ha-channel) and the master dark.

Did anybody got same problems? Where the error could be? If somebody still shot this target with the same camera, maybe let me know the setup of the processes? Or get in touch with me? Sehr gerne auch in deutsch ;-)

Thanks alot my friend for comments and help, will provide original data and infor when needed.

CS!

Jens

[bulrichl]

Do the dark frames match the light frames (exposure time, gain, offset and temperature), how did you make the MasterDark and what were your settings for the image calibration of the light frames?

Bernd

[The_J]

Dear Bernd,

exposure time, gain, offset and temperature are identical. I formely worked with DSS and I normally use identical values for the different file types, as requested. Also in DSS, I cannot remove the Amp-Glow with the Darks.  I used ImageCalibration, ImageIntegration, StarAlignment, CosmeticCorrection. The exact values I have to restore because I stii, don´t know how to safe the workspace. Its beginning at zero every time (sorry, newby errors...). Is there any saved file where I cound recognize the values and processing steps of the images?

Thanks!

Jens

[bulrichl]

Load one of the calibrated light frames and take a look at the FITS header (in the menu: FILE/FITSHeader). The section that begins with "COMMENT     Calibration with PixInsight 1.8.8" contains the settings for ImageCalibration. Take a screen shot and show it.

Bernd

[The_J]

Thanks, Bernd,

I have attached the FITS Header of one Light and the Master dark, maybe that helps to find the problem...

CS!

Jens

[bulrichl]

I am supposing that your MasterDark is the plain integration of the dark frames, and that the flat frames were calibrated with the MasterBias, then integrated to the MasterFlat.

Your settings for the calibration of the light frames were:

-------------------------------------
Section 'Master Bias' enabled
MasterBias selected
Option 'Calibrate' disabled

Section 'Master Dark' enabled
MasterDark selected
Both options 'Calibrate' and 'Optimize' enabled

Section 'Master Flat' enabled
MasterFlat selected
Option 'Calibrate' disabled
-------------------------------------

That means, dark frame optimization was performed. For the calibration of the light frames a Master Bias was used. In the first step of the calibration process, the Master Bias was subtracted both from the target light frame and from the MasterDark. Then the correlation factor k0 for the calibrated MasterDark was determined and the product of k0 and the calibrated MasterDark was subtracted from the difference of the target light frame and the MasterBias. Finally this result was divided by the MasterFlat and multiplied by the mean of the MasterFlat.


Your settings for the calibration of the light frames are OK in section 'Master Flat'. Now please try the following:

-------------------------------------
Section 'Master Bias' disabled

Section 'Master Dark' enabled
MasterDark selected
Both options 'Calibrate' and 'Optimize' disabled

Section 'Master Flat' enabled
MasterFlat selected
Option 'Calibrate' disabled
-------------------------------------

That means, no dark frame optimization will be performed: For the calibration of the light frames no Master Bias is used. In the first step of the calibration process, the MasterDark is subtracted from the target light frame. Then the result is divided by the MasterFlat and multiplied by the mean of the MasterFlat.

Explanation:
Some CMOS sensors show strong "amplifier glow". The IMX183 exhibits particularly strong amp glow. This artifact is not strictly proportional to the exposure time. This is the reason that the amp glow is not completely removed in the calibration step when dark frame optimization is enabled.

Please let me know whether this helps.

Bernd

[The_J]

Dear bernd,

this method helped a large bit. It´s not yet perfect, but the result is much better than before. I have attached 2 images (before and after) for comparison. I´m not sure if I could remove the small rest of amp glow with other settings in calibration or registration. If you have a working "workflow of settings" that would be really helpful.

Thanks alot for your help!

CS!

Jens