Author Topic: Problem with image integration in PI 1.06.09?  (Read 3642 times)

Offline dlp

  • Newcomer
  • Posts: 17
Problem with image integration in PI 1.06.09?
« on: 2010 December 22 06:27:46 »
I am trying to do an integration of bias frames, previously having done a batch convert from TIF* to FITS format.
The process is failing 'Zero or insignificant signal detected: empty image?'. I can open the TIF files and the subsequent FITS files fine in PS CS5 (latter using fits liberator). I never had a problem before with TIF files. Any comments please?

* I have deliberately reduced the original CR2 files to 1600 pixels wide TIF in Lightroom to save processing time - always worked fine before even if it is not an 'ideal' workflow.

David

astropixel

  • Guest
Re: Problem with image integration in PI 1.06.09?
« Reply #1 on: 2010 December 23 15:45:29 »
You could convert the cr2 files using the Batch format conversion script, and then resample the .fit files, prior to integration. That seems a simpler process, keeping the process native to PI.

Offline pfile

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 4729
Re: Problem with image integration in PI 1.06.09?
« Reply #2 on: 2010 December 24 00:50:20 »
that's how i prepare the bias frames to make a master bias frame: CR2->fits using DSLR_RAW set to raw bayer RGB.

you should not convert the bias frames to TIF with lightroom, it's going to make the linear data nonlinear as part of it's debayer process... the only way to properly calibrate DSLR images is to work in bayerRGB format until calibration is finished. at that point you can use the BatchDebayer script to interpolate the images, and then register and stack the debayered images.


Offline dlp

  • Newcomer
  • Posts: 17
Re: Problem with image integration in PI 1.06.09?
« Reply #3 on: 2010 December 26 10:57:20 »
OK, thanks Guys. I'll completely convert to using raw in the whole process I think, despite the processing times. Just ordered 8Gb RAM - think PI will eat that!