Author Topic: Which pixel rejection algorithm  (Read 5591 times)

astropixel

  • Guest
Which pixel rejection algorithm
« on: 2011 June 29 22:38:50 »
I have 30 dark frames to be integrated and interested in what pixel rejection algorithm, if at all, would be a preference for other users. Currently, I am using Windsorized with high and low rejection at 4 and 4.

Offline DarrenS

  • Newcomer
  • Posts: 35
Re: Which pixel rejection algorithm
« Reply #1 on: 2011 June 30 06:42:31 »
My understanding is..

If you want the best results it’s a case of trying different algorithms and checking the snr to pick the best algorithm

Snr values are displayed in the console at the end of each stack

After selecting the best algorithm it’s then a case of enabling pixel rejection and tweaking the settings until you reject any unwanted artifacts while keeping the snr as high as possible

Btw to speed this process up > stack with NO pixel rejection > select a preview > use the preview to fine tune your parameters

The top article on this page
http://astrosurf.com/jordigallego/articles.html
Is full of excellent information and well worth reading.

hope this is of some help
Darren

astropixel

  • Guest
Re: Which pixel rejection algorithm
« Reply #2 on: 2011 June 30 12:27:48 »
Thanks Darren. Much appreciated. Will follow up.

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 7111
    • http://pixinsight.com/
Re: Which pixel rejection algorithm
« Reply #3 on: 2011 June 30 12:35:15 »
Hi John (oops!) Rowland,

I agree with Darren's advice, and I also recommend you read Jordi's article on image integration algorithms. It is one of the most thorough and serious practical works I've seen on the subject.

With 30 dark frames I'd say Winsorized sigma clipping is your best option. If they were light images I'd encourage you to use linear fit clipping, but for darks it's probably useless. Both algorithms should work nicely with 30 images anyway.
« Last Edit: 2011 June 30 12:59:29 by Juan Conejero »
Juan Conejero
PixInsight Development Team
http://pixinsight.com/

Offline Sean

  • PixInsight Addict
  • ***
  • Posts: 144
    • My Personal Website
Re: Which pixel rejection algorithm
« Reply #4 on: 2011 June 30 19:20:43 »
Juan,

I went back and revisited Jordi's presentation, which now makes a lot more sense. However, in doing some experimenting with a stack of 20 calibrated frames, I noticed that the Avg SNR Increment is actually a bit higher for the Windsorized Sigma and Linear Fit, compared to the Average stack with no pixel rejection. I thought that the SNR should be highest with Average/No Rejection, and that the various rejection algorithms could then be tested against their approach to this best value. Am I missing something?

Thanks,

Sean

astropixel

  • Guest
Re: Which pixel rejection algorithm
« Reply #5 on: 2011 July 01 13:03:39 »
Thanks Juan, will do. Interestingly, linear clipping is my choice for lights and Windsorized for darks, but frankly, I was just following the tool tips and Vicent's examples.