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

DeepSky Stacker - Bad Frames

  • Please log in to reply
12 replies to this topic

#1 TheBrochan

TheBrochan

    Explorer 1

  • -----
  • topic starter
  • Posts: 74
  • Joined: 30 Jul 2020

Posted 25 November 2021 - 04:53 PM

I was searching around and maybe I'm not looking for the right thing, its probably really simple and I'm missing it.

 

I had some bad frames during my last session so DSS threw them out. I'd like to permanently delete from from my HD but i'm not seeing where it lists the frames that were rejected. I noticed a few has scores 100 and under while the rest had 400+, would the ones 100 and lower be the ones that got rejected?



#2 Oort Cloud

Oort Cloud

    Apollo

  • -----
  • Posts: 1,029
  • Joined: 19 Nov 2020
  • Loc: New Jersey, USA

Posted 25 November 2021 - 05:01 PM

Yes, they'll also likely have very low star counts, which is why they don't get included.

#3 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 6,140
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 25 November 2021 - 05:12 PM

Depends what the stacking settings were, but probably. If you set it to stack the best 90% then the 10% with lowest scores wont have been included.



#4 Oort Cloud

Oort Cloud

    Apollo

  • -----
  • Posts: 1,029
  • Joined: 19 Nov 2020
  • Loc: New Jersey, USA

Posted 25 November 2021 - 07:42 PM

That's true too. I was assuming you had it set to 100%

#5 Mike in Rancho

Mike in Rancho

    Surveyor 1

  • -----
  • Posts: 1,624
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 25 November 2021 - 09:21 PM

Okay, assuming you are not setting a percentage to stack, and that you didn't use the threshold button for adding checks to boxes, I'm moderately sure that if it has a score, it will be used.  No matter how lousy that score might be.  tongue2.gif

 

Just bring up the same list, if you save it, or re-load all those same subs into lights.  Their scores from the most recent registration should be loaded with them.

 

Another way is to run compute offsets.  Those subs that will stack will have their dX, dY, and angle columns filled in.  The rejects will be blank or maybe NC.

 

Finally, in the lights folder you'll also find DSS' stackinfo.txt file, which will explain what it included.  A far more tedious way of going about it, I would think.


  • TheBrochan likes this

#6 Oort Cloud

Oort Cloud

    Apollo

  • -----
  • Posts: 1,029
  • Joined: 19 Nov 2020
  • Loc: New Jersey, USA

Posted 26 November 2021 - 01:54 AM

Just  use Siril :)


  • lock042 and TheBrochan like this

#7 TheBrochan

TheBrochan

    Explorer 1

  • -----
  • topic starter
  • Posts: 74
  • Joined: 30 Jul 2020

Posted 26 November 2021 - 10:01 AM

Thanks for the responses all!

 

Currently have it set to 95%

 

As an example, I shot roughly 2.5 hours of data. After DSS registered and stacked I was left with around 1.3 hours of data. I figured theres no point in keeping the rejected data around and letting it take up space on my hard drives which was why I wanted to get rid of them. It seems DSS doesn't keep a list or log of the frames that were rejected.


Edited by TheBrochan, 26 November 2021 - 10:01 AM.


#8 Mike in Rancho

Mike in Rancho

    Surveyor 1

  • -----
  • Posts: 1,624
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 26 November 2021 - 02:34 PM

To be honest, I would make sure your files are truly bad for some reason before trashing them.  Find out why DSS is rejecting them (and I presume you mean any rejects beyond the 5% you are already rejecting off the top by choice?).

 

All programs will come up with their own rankings or scores, and they are not identical.  What DSS considers the worst sub could be up the list a bit more in ASTAP.

 

I always use 100% for DSO.  For planets and something like AS3, sure, hundreds or thousands of frames it's best to set up the software to help you out.  But often enough, especially if I tagged and removed obviously bad frames during acquisition (wind gust or guiding hiccup), DSS' lowest-scored sub is still definitely worth including.

 

You aren't experiencing any kind of significant drift or field shift, are you?  Because DSS will also reject due to commonality of stars, which can be dependent upon the reference frame selected.


  • Oort Cloud likes this

#9 Oort Cloud

Oort Cloud

    Apollo

  • -----
  • Posts: 1,029
  • Joined: 19 Nov 2020
  • Loc: New Jersey, USA

Posted 26 November 2021 - 03:31 PM

All good stuff above.  It is true, eventually you'll just get to a point where all the subs are good, usually because you're deleting the bad ones as soon as the shutter closes.  In that case, there will still be one with the lowest score, so you don't want to discard something just because of what some program says about it.  But we don't always have time to watch every sub's autopreview to determine if it should stay or go, and sometimes we have bright stars that require us to shoot an enormous amount of really short subs.  So, if I'm not doing a visual analysis of each light frame prior to stacking, generally what I look for is subs that are outliers; rather than looking for a specific FHWM or roundness or background or number of stars, I look for subs that deviate from the rest by a significant margin in any of those categories.  Those are usually the ones that wind up on the chopping block. FWHM for the set is 2-3, but 4 subs have over 5?  Toss em.  Roundness for the set is .8 or better, but 2 subs are at .6?  Toss em.  If you find that you have a lot of variation in your subs, work on figuring out why.  There will always be some outliers, but if you're repeatedly getting bad subs mixed with good ones, then that's a problem.  And the rest of the process all gets much easier when you start with good data, so if you're still new to all of this, it helps more than I can describe to remove bad subs from your stack.  So take the time to visually check all of your lights with a stretch applied, looking for things like focus (FWHM), tracking (FWHM/roundness), and high clouds (background), until you get familiar with what ranges are normal for those figures.



#10 TheBrochan

TheBrochan

    Explorer 1

  • -----
  • topic starter
  • Posts: 74
  • Joined: 30 Jul 2020

Posted 26 November 2021 - 06:54 PM

Thanks again for the responses. I did check these frames before discarding. Wind definitely had an effect on them. Also the area of the sky I was shooting has planes flying through there especially earlier in the evening. Usually by 8pm it slows down.

Star count was very low as well when comparing to other frames that I shot. I had a feeling that some wouldn't be good due to the wind gusts that evening but I did get a few that turned out great.

Next set of data I stack I will try with 100% and 95% and see if I notice any differences.

#11 starblue

starblue

    Viking 1

  • -----
  • Posts: 621
  • Joined: 01 Apr 2015
  • Loc: St. George, Utah, USA

Posted 26 November 2021 - 11:56 PM

You can get DSS to save a log of what it did.

 

Go to Settings->Stacking Settings, then in the "Output" tab->"Output File settings" area, checkmark the box "Create HTML description file". The resulting file with a .html suffix will tell you exactly what files DSS stacked. It's worth examining the other options there too--you might want some of them, like the directory where to write the output files.



#12 TheBrochan

TheBrochan

    Explorer 1

  • -----
  • topic starter
  • Posts: 74
  • Joined: 30 Jul 2020

Posted 27 November 2021 - 12:02 PM

You can get DSS to save a log of what it did.

Go to Settings->Stacking Settings, then in the "Output" tab->"Output File settings" area, checkmark the box "Create HTML description file". The resulting file with a .html suffix will tell you exactly what files DSS stacked. It's worth examining the other options there too--you might want some of them, like the directory where to write the output files.


Excellent! Thanks very much. Will definitely give this a look.

#13 rj144

rj144

    Viking 1

  • -----
  • Posts: 831
  • Joined: 31 Oct 2020

Posted 27 November 2021 - 12:36 PM

If it is 100% stack, it threw out the frames it couldn't align which are listed as NC in the offsets.


  • happylimpet 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