View Single Post
  #21  
Old 27-10-2013, 09:05 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by gregbradley View Post
Speaking of that, what rejection parameters have you found workable? I tend to only use hot and cold pixel rejection and have noticed that helps. I have not noticed the other rejections to help at all unless I am simply using the wrong settings.

Greg.
Greg,

I use CosmeticCorrection for the hot and cold pixels and then I experiment with a couple of different rejection algorithms depending on how many subs I have. For a handful of subs, Percentile Clipping usually produces the best results. A few more and you're in Sigma Clipping country. When you get up to 10 or so that's Winsorized Sigma Clipping and then 15+ is usually good for Linear Fit Clipping. ImageIntegration spits out some SNR data so you can compare how well each algorithm works. I start with an integration with no rejection which will give me an upper limit for the SNR. Then I play with the algorithms and parameters to see how close I can get to that SNR and still get rid of any satellite trails, cosmic rays, etc. It can be a slow process but will result in a close to optimal SNR.

Cheers,
Rick.
Reply With Quote