Flats not being applied.

J

jjhughes53

Well-known member
I took 60 67second exposures of M101@ 500 Gain. Took the same number of darks at the same exposures and a made a master dark. . I used a superbias. I also took 60 flats at .032 second and 500 gain and used image calibration to make both a master dark and a master flat Using image calibration. The master flat looked normal with dust spots and vignetting. But when I calibrated the master flat did not appear to be applied to the lights as the dust spots were still there. I followed the directions off the pixinsight 1.8 calibrating and stacking video and tried again but it does not appear the flat is applied to the lights. Does anyone have a suggestion?
 
Dark frames shall not be pre-calibrated with the MasterBias or superbias. Simply integrate the dark frames to the MasterDark. Use this new MasterDark for the calibration of your light frames.

Then calibrate the light frames with one of the following settings, depending whether you want to apply dark frame optimization or not:

1) No dark frame optimization:
Disable section 'Master Bias'.
Enable section 'Master Dark', select the MasterDark, disable both options 'Calibrate' and 'Optimize'
Enable section 'Master Flat', select the MasterFlat, disable the option 'Calibrate'

2) With dark frame optimization:
Enable section 'Master Bias', select the MasterBias, disable the option 'Calibrate'
Enable section 'Master Dark', select the MasterDark, enable both options 'Calibrate' and 'Optimize'
Enable section 'Master Flat', select the MasterFlat, disable the option 'Calibrate'

Check whether the calibrated light frames are clipped (histogram and statistics). If this is the case, apply an output pedestal in the calibration of the light frames in order to avoid clipping.

The application of dark frame optimization is not recommendable if your camera shows strong "amplifier glow".

Bernd
 
Hey Bernd,

Not being a PI expert, if I look at the 'Statistics' of one of my calibrated Lights (what needs to be checked/unchecked in Statistics ?), how do you confirm that the image has actually been clipped, further, if Statistics proves this to be the case, how would you decide what figure to use as the Pedestal amount (I assume you would use 'Literal Value' for Mode) ? ? ?

Further, is it also possible to clip the Flats when calibrating them too requiring a similar action.

Last one, If the lights were clipped at calibration time and a PI user just integrated them anyway without redoing them with a Pedestal as you suggest, what would you see in the final master light integration....100% black pixels dotted about the image for example ?

Many Thanks
Paul
 
Hi Paul,

clipping can easily be detected by viewing the histogram at higher horizontal zoom (e.g. 50 - 100). If the peak on the left side is sharply cut at an intensity of zero, clipping is indicated.

In 'Statistics' of a calibrated light frame you can read out the extent of clipping quantitatively. Please take a look at the value of 'count (%)'. It should be very close to 100 %. The rest to 100 % is the fraction of clipped pixels. In my subjective view, count (%) should be >= 99.99 %, i.e. the fraction of clipped pixels should be < 0.01 %. If this condition is not met, the application of an output pedestal in image calibration is advisable.

For applying an output pedestal, you need not alter anything in the section 'Pedestal'. An output pedestal can be set in the section 'Output File' of the ImageCalibration process. The value shall be set in order that no significant clipping is detected in the calibrated light frames. The range of input values for the output pedestal is 0 - 1000 DN. It doesn't hurt when the chosen value is a little bit too large. Normally a value in the range of 100 - 200 DN is sufficient. When an output pedestal is applied, the second to last step of the ImageCalibration process is the addition of this pedestal to all pixel intensities, before (as the last step) truncation to the range [0,1] is performed.

One important point: IF you apply an output pedestal in image calibration, the option 'Subtract pedestals' in the image integration later in the workflow MUST be disabled. Otherwise the first step in the ImageIntegration process will be the subtraction of the pedestal, producing the same result as if no pedestal was applied before.

Clipping will not take place during the calibration of the flat frames, unless your flat frames are severely underexposed.

I think clipping is easier detectable in the calibrated light frames. Usually it will suffice when you spot-check some of the calibrated light frames (histogram and statistics).

Bernd
 
bulrichl said:
Dark frames shall not be pre-calibrated with the MasterBias or superbias. Simply integrate the dark frames to the MasterDark. Use this new MasterDark for the calibration of your light frames.

Then calibrate the light frames with one of the following settings, depending whether you want to apply dark frame optimization or not:

1) No dark frame optimization:
Disable section 'Master Bias'.
Enable section 'Master Dark', select the MasterDark, disable both options 'Calibrate' and 'Optimize'
Enable section 'Master Flat', select the MasterFlat, disable the option 'Calibrate'

2) With dark frame optimization:
Enable section 'Master Bias', select the MasterBias, disable the option 'Calibrate'
Enable section 'Master Dark', select the MasterDark, enable both options 'Calibrate' and 'Optimize'
Enable section 'Master Flat', select the MasterFlat, disable the option 'Calibrate'

Check whether the calibrated light frames are clipped (histogram and statistics). If this is the case, apply an output pedestal in the calibration of the light frames in order to avoid clipping.

The application of dark frame optimization is not recommendable if your camera shows strong "amplifier glow".

Bernd
Click to expand...
Thank you for your suggestion. My darks do not have bias calibrated into them. But I tried to calibrate the lights both with and without dark frame optimization and calibration enabled and disabled but still have the dust spots on the calibrated lights. Is there anything else I can try or should just delete them and start over?
 
bulrichl said:
How did you calibrate the flat frames?

Bernd
Click to expand...
I took 60 flat frames at .032 seconds and 500 G. The historgram showed a peak between 40 and 60. I followed the pixinsight 1.8 calibrate and stacking video on you tube. I am not sure who made it. I pulled up image calibration and added the flats images. Created an output directory. Selected my master superbias for master bias. Then selected my master dark That I took form 60 67 second 500G darks at -10C. I did choose calibrate. On Master bias as I am using a ZWO 294 pro cooled camera. I chose calibrate and Optimize on the Master Dark panel.
 
the dust is probably on the sensor or any reducer. dust on the corrector plate or camera lens is too far away from the sensor to be seen - the shadow cast from dust on the corrector plate is bigger than the sensor.

rob
 
pfile said:
the dust is probably on the sensor or any reducer. dust on the corrector plate or camera lens is too far away from the sensor to be seen - the shadow cast from dust on the corrector plate is bigger than the sensor.

rob
Click to expand...
Thanks I am not using a reducer at this point. I will check the camera sensor again. What is really frustrating is I was able to process images fine last week and now for some reason I am plagued with dust spots. I can’t figure out why the flats can not correct them.
 
For further assistence I would need a light, dark, bias and flat frame (in FITS format) and the MasterDark, MasterBias and MasterFlat in XISF format. If you upload these data to a filehoster and post the link here, I will take a look at it.

Bernd
 
bulrichl said:
For further assistence I would need a light, dark, bias and flat frame (in FITS format) and the MasterDark, MasterBias and MasterFlat in XISF format. If you upload these data to a filehoster and post the link here, I will take a look at it.

Bernd
Click to expand...
Thanks I got out for a few hours last night and shot a whole new set of lights followed up by a new set of darks and flats this morning. They processed as usual without an issue. I don’t know if I had a bad setting on the other set that prevented it from processing but I seem to be where I need to be now. Thank You for your kind assistance.
 
I am glad that now it works for you. On the other hand this outcome is quite unsatisfactory, because nobody can learn anything from this thread.

Bernd
 
bulrichl said:
I am glad that now it works for you. On the other hand this outcome is quite unsatisfactory, because nobody can learn anything from this thread.

Bernd
Click to expand...
Well, you sure are correct about that. It looked like I was just about to find the solution to my problem there for a little while. And then that happened......Regarding calibration process for MD, MF and ML, I think I have now seen every combination of checked or unchecked Calibrate and/or Optimize boxes as being the correct method to use within the ImageCalibration Process. Here is one them that didn't work for me....(Ignorance is the root of all frustration.......Color me frustrated....)


1682641035482.png
 
Last edited:
DustinGass said:
Here is one them that didn't work for me....
Click to expand...
Since you didn't describe WHAT didn't work for you, it is not a good idea to resurrect a 3 year old thread. You better open a new thread and describe your issue in detail.

If the cause is not obvious it might be necessary to upload the data that I specified in post #11 of this thread.

Bernd
 
You must log in or register to reply here.
Back
Top