seqsubsky fail in batch scripts, fits excluded

Software Siril
1

I use a script that includes the seqsubsky command for background extraction (I modified the OSC_prprocessing script).
…

Convert Light Frames to .fit files

cd lights
convert light -out=…/process
cd …/process

Pre-process Light Frames

preprocess light -dark=dark_stacked -flat=pp_flat_stacked -cfa -equalize_cfa -debayer

Extract linear gradient

seqsubsky pp_light 1

Align lights

register bkg_pp_light
…

So far, no problems. With the latest fits files this error appears:

09:51:38: # Extract linear gradient
09:51:38: Running command: seqsubsky
09:51:38: Background Extraction: with the current memory and thread limits, up to 4 thread(s) can be used
09:51:38: Background Extraction: processing…
09:51:38: Reading FITS: file pp_light_00001.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Reading FITS: file pp_light_00005.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Reading FITS: file pp_light_00004.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Reading FITS: file pp_light_00003.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Reading FITS: file pp_light_00002.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Reading FITS: file pp_light_00006.fit, 3 layer(s), 4144x2822 pixels
09:51:38: Sequence processing partially succeeded, with 6 images that failed and that were temporarily excluded from the sequence.
09:51:38: Execution time: 640.61 ms.
09:51:38: # Align lights
09:51:38: Running command: register
09:51:38: Loading sequence `bkg_pp_light’ failed.
09:51:38: Error in line 65: ‘register’.
09:51:38: Exiting batch processing.

The camera is an ASI294 MC Pro, Siril 1.0.4.
Is it possible that the error depends on the software used during the imaging session? The strange thing is that the error first occurred with the files acquired with N.I.N.A. . Until now with Sharpcap, however, no error. No error even if I rework old Sharpcap fits.
Only N.I.N.A. fits are rejected.

2

Hello, this issue was fixed last week, but for 1.2 version. It is just a lack of reporting: background extraction is not supported for images with many pixels of negative value.
The real problem (I hope) here is that calibration is not done with the correct files. Check your masters or raw calibration files, check the raw images, something must be wrong.
Thanks

3

Thanks Vincent
For the calibration files: biases and darks are the usual ones I use (from previous sessions), I took the flats specifically for this session. They look fine to me. What other reason could there be for “many pixels of negative value”? I wanted to use N.I.N.A. to quickly test the feature for making mosaics; as a quick test session, could it be that I used exposure times that were too short?

4

Using these three types of files is fine, but it depends how you use them and if they share the same required settings.
If you use the default scripts then that would be an acquisition settings problem. If not, maybe you’re not calibrating the darks and subtracting both bias and dark to lights?
In acquisition settings, make sure the lights, bias and darks were made with the same software and offset parameter, that darks and lights had the same exposure and temperature, and so on.

5

Hi Vincent,
I did the darks with N.I.N.A. and the problem was solved.
Great software.
Thanks again.