Author Topic: ImageIntegration on Two Images  (Read 678 times)

Offline akulapanam

  • Member
  • *
  • Posts: 59
    • View Profile
ImageIntegration on Two Images
« on: 2017 October 23 21:28:53 »
I was curious why ImageIntegration doesn't allow you to do a noise weighted integration of only two images.  I have a situation where I'm shooting SH2 187 and I want to use some ha frames I took to enhance Lum because neither stands on its own.  I could use PixelMath but I would like to use a noise evaluation based weight but unfortunately it appears that the imageintegration process requires at least 3 images.

Offline RickS

  • PTeam Member
  • PixInsight Jedi
  • *****
  • Posts: 1295
    • View Profile
Re: ImageIntegration on Two Images
« Reply #1 on: 2017 October 23 22:48:17 »
I presume the reason why ImageIntegration expects at least three images is that you can't do rejection with any fewer.

If you want to do the noise-weighted averaging without rejection you can "trick" ImageIntegration: just add the same files twice if you are dealing with FITS files.  With XISF files there is some file locking that prevents this (you get an error when you run the process) but you can just make copies and add those.

Cheers,
Rick.

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 6782
    • View Profile
    • http://pixinsight.com/
Re: ImageIntegration on Two Images
« Reply #2 on: 2017 October 24 04:53:13 »
Quote
With XISF files there is some file locking that prevents this

Please note that this bug (wrong file open share mode on Windows) was already fixed. This does not happen in version 1.8.5 of PixInsight.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/

Offline RickS

  • PTeam Member
  • PixInsight Jedi
  • *****
  • Posts: 1295
    • View Profile
Re: ImageIntegration on Two Images
« Reply #3 on: 2017 October 24 13:48:21 »
Please note that this bug (wrong file open share mode on Windows) was already fixed. This does not happen in version 1.8.5 of PixInsight.

Great, thanks Juan!  I hadn't tried this in a while.