Should image calibration with dark fram remove hot pixels?

Hi,
As far as I understand it, ImageCalibration should remove hot pixels, if I provide a dark frame. It did not. I do understand, that I can fix that with CosmeticCorrections, but I am not sure if I am doing something wrong in step 1. Please find attached the following image: top right is my master dark (you can see two hot pixels). Top left is my raw image, again you can see the same two hot pixel. I ran ImageCalibration and the result is the picture at the bottom. The two hot pixels are still there. Any idea what I did wrong?
Thanks, Rainer
 

Attachments

  • HotPixel.jpg
    HotPixel.jpg
    264.5 KB · Views: 102
Reasons for image calibration producing results with remaining hot pixels can be different, examples are:

- dark frames do not match the light frames (temperature, gain, offset, exposure time),
- wrong settings were applied for image calibration (e.g. dark frame optimization is not recommended for certain cameras).

So this question cannot be answered without further information:

- how were the dark frames prepared,
- do they match the light frames in each aspect,
- which camera is used,
- what were the settings in image calibration.

Bernd
 
Hi Bernd,

Thank you for your reply! Here are the answers to your questions:

- how were the dark frames prepared,
The camera was shut, the cooling was set to the same temperatur (-10°C), the exposure time was the same (60 sec), master bias was substracted
- do they match the light frames in each aspect,
As far as I can say, yes.
- which camera is used,
Omegon Camera veTEC 16000 C Color, using the CMOS Chip (Panasonic MN34230), see Omegon Camera veTEC 16000 C Color (astroshop.eu)
- what were the settings in image calibration.
Default, although I also tried to set the Optimization threshold to 0, same result

Thanks, Rainer
 
And you used Optimize Darks?
Defaults are an illusion (just saying).
Dark Optimization will not characterize hot pixels in general. A straightup subtraction of a matching dark (binning, time, temperature) is the best way to see if your dark is characterizing your data and you are on the right foot.

-adam
 
Hi Adam,

Thanks for your reply. I am not exactly sure, if I understood you.

1.) What do you mean by optimized darks? I created a master dark as described in various books and on the internet with PixInsight.
2.) I thought the Image Calibration will exactly that, substract my master dark from the lights. Am I wrong there?
3.) I used PixelMath and substracted the master dark from my light (left image). The result (right image) has nearly all hotpixels removed.

Cheers, Rainer
 

Attachments

  • Subsctract.jpg
    Subsctract.jpg
    661.9 KB · Views: 67
Rainer:

Brilliant! You proved my point!!
The answer to #1 is proved by #3.
If you get a chance..please check out my series on WBPP. I explain *all* about it!

If you have Optimize Dark checked, it can certainly lead to the behavior you are seeing with regards to hot pixels.
-adam
 
Hi Rainer,

- how were the dark frames prepared,
The camera was shut, the cooling was set to the same temperatur (-10°C), the exposure time was the same (60 sec), master bias was substracted
- do they match the light frames in each aspect,
As far as I can say, yes.
- which camera is used,
Omegon Camera veTEC 16000 C Color, using the CMOS Chip (Panasonic MN34230), see Omegon Camera veTEC 16000 C Color (astroshop.eu)
- what were the settings in image calibration.
Default, although I also tried to set the Optimization threshold to 0, same result
Three points:
1. You should not calibrate the dark frames with a MasterBias. Simply integrate the dark frames to the MasterDark.
2. Don't use the MasterBias in the calibration of the light frames.
3. Disable 'Dark frame optimization' in ImageCalibration with this camera.

Bernd
 
Thanks Bernd! One remark on Bias frames: I am not sure if I capture them correctly, but basically my Bias frames are more or less emtpy.
 
The Panasonic MN34230 sensor is notorious for having an inconsistent bias level when exposure times < 0.2 s are used. With a camera utilizing this sensor, it is not advisable to use bias frames at all (i.e. frames that are captured with the sensor in complete darkness, at the shortest exposure time that the camera can provide which is achieved by setting an exposure time of 0 s).

So bias frames will not be needed in the calibration of the light frames. However you will need calibration frames for the calibration of the flat frames. For your camera it is advisable to capture flat-darks (i.e. dark frames with an exposure time matching the flat frames), prepare a MasterFlatDark and calibrate the individual flat frames with it..

Conclusion:
For your camera, the following calibration frames are needed: dark frames, flat frames and flat-darks.
Dark frames and flat-darks have to be integrated to MasterDark and MasterFlatDark respectively.
The individual flat frames have to be calibrated with the MasterFlatDark.
The light frames have to be calibrated with the MasterDark (NOT using 'Dark frame optimization') and the MasterFlat.

If following this approach there are remaining hot pixels in the calibrated light frames, then it is recommendable to use CosmeticCorrection in order to remove them.

Bernd
 
Back
Top