PixInsight Forum

PixInsight => General => Topic started by: Andrei Ioda on 2016 June 15 00:21:16

Title: Rejection issue
Post by: Andrei Ioda on 2016 June 15 00:21:16
Hi,

I have issue with rejection of satellite tracks. I tried any settings (Rejection algorithm, Normalization, Sigma high and etc.) in the ImageIntergration but didn't get result. I use PixInsight Core 01.08.04.1195 Ripley (x64) on OSX 10.11.5.

You can see the result on the attached screenshot and can download my original images for integration from here (https://www.dropbox.com/sh/9s68nfme1kah8k8/AAB1C4cY0plwCxv-sLvhwmYta?dl=0).
Title: Re: Rejection issue
Post by: Carlos Milovic on 2016 June 15 10:17:24
You'll have to further reduce the sigma high parameter to reject the trails.
Title: Re: Rejection issue
Post by: RickS on 2016 June 15 16:55:08
I have found, especially with small stacks, that rejection settings strong enough to clean up satellite trails can result in rejection of good data in the rest of the image.  My solution is to zap the satellite trails in the individual subs using PixelMath by drawing a white (value 1.0) line over the trail.  These lines reject very easily with weaker rejection settings.  It's not fun to do but I find it worth the effort.

Cheers,
Rick.
Title: Re: Rejection issue
Post by: Andrei Ioda on 2016 June 15 23:25:37
I have found, especially with small stacks, that rejection settings strong enough to clean up satellite trails can result in rejection of good data in the rest of the image.  My solution is to zap the satellite trails in the individual subs using PixelMath by drawing a white (value 1.0) line over the trail.  These lines reject very easily with weaker rejection settings.  It's not fun to do but I find it worth the effort.

Cheers,
Rick.

Thanks Rick. I will try.
Title: Re: Rejection issue
Post by: RickS on 2016 June 16 01:46:08
Thanks Rick. I will try.

Good luck, Andrei.  I use a PixelMath expression "iif(d2seg(x1,y1,x2,y2)<r,1,$T)" and define the symbols x1,y1 and x2,y2 as the endpoints of the satellite trail and r as the thickness of the line.

Cheers,
Rick.
Title: Re: Rejection issue
Post by: msmythers on 2016 June 16 04:54:48
Rick

Thanks for sharing and explaining. I had never thought of that type of method for repairing an integration. Now I have some old images to work on. ;)

Andrei  You are going to like the results.


Mike
Title: Re: Rejection issue
Post by: jkmorse on 2016 June 16 07:12:26
Rick,

Mind if I include that in my next version of my workbook?  Great solution.

Best,

Jim
Title: Re: Rejection issue
Post by: RickS on 2016 June 16 13:50:53
Mind if I include that in my next version of my workbook?  Great solution.

Please do include it, Jim!
Title: Re: Rejection issue
Post by: Andrei Ioda on 2016 June 16 23:20:06
Rick, Mike.

Thanks. I tried this method and it works. As the result of my work of M63 - http://www.astrobin.com/252475/C/
Title: Re: Rejection issue
Post by: msmythers on 2016 June 16 23:28:25
Andrei

The image turned out great. Well worth the extra effort to remove the tracks.


Mike
Title: Re: Rejection issue
Post by: Andrei Ioda on 2016 June 16 23:31:04
Andrei

The image turned out great. Well worth the extra effort to remove the tracks.


Mike

Yes. Thanks Rick.