Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Remove satelitetrails in APP

  • Please log in to reply
9 replies to this topic

#1 taipan

taipan

    Lift Off

  • -----
  • topic starter
  • Posts: 18
  • Joined: 01 Sep 2019

Posted 23 September 2019 - 04:49 AM

Got the Trail version of APP to se if that program is anything for me. Can anybody help me with witch settings i should use to remove satelitetrails?

And if i use exposeurs from different occasions when i stack, something i should think about or will the program align the pictures automatic?

 

Tried the program yesterday and used the standard settings.



#2 leveye

leveye

    Aurora

  • ***--
  • Posts: 4812
  • Joined: 06 Feb 2013
  • Loc: Central Oregon Coast

Posted 23 September 2019 - 05:43 AM

Got the Trail version of APP to se if that program is anything for me. Can anybody help me with witch settings i should use to remove satelitetrails?

And if i use exposeurs from different occasions when i stack, something i should think about or will the program align the pictures automatic?

 

Tried the program yesterday and used the standard settings.

Don't use the exposures with trails. Problem solved.for that issue.


  • scottmm2012 likes this

#3 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3995
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 23 September 2019 - 05:46 AM

Use kappa-sigma rejection. I assume APP does that, everything else does!

 

The satellite trails will just get thrown out, though the rest of the sub will be used - much better than just omitting the offending subs.



#4 barrett_flansburg

barrett_flansburg

    Explorer 1

  • -----
  • Posts: 81
  • Joined: 30 Dec 2014

Posted 23 September 2019 - 01:12 PM

I agree that the best option is to exclude any known bad subs from the integration. However, if you have dozens or hundreds of subs it may not be practical to examine every one of them to look for flaws like airplane or satellite trails. Regardless, you can select one of the many filters in the drop down box for “outlier rejection” that appears midway down on the INTEGRATE tab. I use the “MAD winsor clip” filter option with the default 3.0 value for kappa. If you hover your mouse over the filter drop down box a description of all of the options will appear. 



#5 physics911

physics911

    Ranger 4

  • *****
  • Posts: 362
  • Joined: 28 Feb 2006
  • Loc: North Georgia, USA

Posted 23 September 2019 - 07:13 PM

Select sigma clipping if you have ~<25 subs, windsor clipping if you have ~>25 subs

 

 Capture.JPG



#6 f300v10

f300v10

    Ranger 4

  • -----
  • Posts: 302
  • Joined: 16 Sep 2018
  • Loc: Marietta, GA

Posted 23 September 2019 - 09:29 PM

I get great results using LN MAD winsor clip in APP. 



#7 eshy76

eshy76

    Sputnik

  • *****
  • Posts: 45
  • Joined: 21 Jun 2018
  • Loc: London, UK

Posted 24 September 2019 - 02:42 AM

Another suggestion which I wish I thought of a few nights ago...if you only have 20 or less subs, you can use "median" instead of "average" for the integration type. Medians will use something like the middle pixel values across all subs, whereas average will use everything, even satellite trails and then try to average it away. If you have 100s of subs, the latter approach is probably fine, but with few subs, there is less to average.

Definitely also use the rejection algorithms also mentioned above.

In my case I also had trails in my integration and just binned the sub in question...but now wish I had tried the above first!

#8 taipan

taipan

    Lift Off

  • -----
  • topic starter
  • Posts: 18
  • Joined: 01 Sep 2019

Posted 24 September 2019 - 03:29 AM

Tried the Mad Win and it worked. Great, now i know witch settings to play with and se witch is working best for me :)

Tnx guys!


  • happylimpet likes this

#9 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3995
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 24 September 2019 - 04:38 AM

Another suggestion which I wish I thought of a few nights ago...if you only have 20 or less subs, you can use "median" instead of "average" for the integration type. Medians will use something like the middle pixel values across all subs, whereas average will use everything, even satellite trails and then try to average it away. If you have 100s of subs, the latter approach is probably fine, but with few subs, there is less to average.

Definitely also use the rejection algorithms also mentioned above.

In my case I also had trails in my integration and just binned the sub in question...but now wish I had tried the above first!

Median is a very poor estimator though, as its tied to using one of the absolute values in the sample, throwing away precision. Much better to use kappa-sigma or winsor as described above.



#10 TimN

TimN

    Gemini

  • *****
  • Posts: 3142
  • Joined: 20 Apr 2008
  • Loc: Ontario, Canada

Posted 24 September 2019 - 07:13 AM

If you have lots of images and don't want to manually choose which ones to use, you should not only follow the above suggestions, but during the integration phase DON'T use the default 'equal' which gives equal weighting to each image. Choose 'quality' which uses the quality ratings to weight each image.


  • f300v10 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics