›

Preselection of sub exposures

Jim Smith
in PixInsight
My Seestar 50 captured 361 x 30s fits files of M81. Looking through them I can see that some have lighter backgrounds (I started capturing a little before it was truly dark) and some have less than perfectly round stars (tracking errors of some sort?). However, most of them look good. 

I assume it would be a good idea to omit the obviously poorer ones from processing to save time and give a better final result. If so, what is the best way to remove the unwanted subs?

Thanks, Jim

Comments

  • BlockHeadBlockHead
    Hmm... a couple of comments.

    In my Fundamentals course (soon to be reorganized)- I explain what constitutes poorer frames.
    Differences in background does not mean a poor image (necessarily).

    None perfect round stars... is not necessarily a problem if the number of them are sufficiently few. The combination of rejection and weighting will still let these frames work for you for what the worth in terms of contributing to the stack.

    Yes, if the frames are truly bad you can omit them. The easiest way (brute force) is while you are blinking through them. You can use the blink tool. There is a button to move files to another place on your disk- I called this a ToBeDeleted folder (which I later do delete). There are some other ways as well. I demonstrate this in my Blink tutorial. I can't remember..I thought I did this in FastTrack.

    -the Blockhead
  • Jim Smith
    I thought there might be some way of getting PixInsight to analyse all of the frames and show me, for example, which ones are the ones with the worst shaped stars and the let me exclude them from further processing.

    If not, I suppose I can select them myself using Blink, which you did cover in the FastTrack course.
  • BlockHeadBlockHead
    You can. There is a different way. You can use SubFrameSelector. I also have a series of videos that explain this method as well.

    However, as I explain in some videos, I think that people have a false sense of "security" when utilizing this methodology and it comes with yet another complex thing that isn't really worth it anymore. With the weighting schemes that are in place right now- you effectively *have* an automated exclusion method (because those bad frames will not be given significant weight). I explain how these new weighting schemes work in my courses. This means you do not need to do a separate step- other than removing the absolute worse frames just to save computational time and the not have them interfere with other processes (for example, if the star quality is so poor, StarAlignment will not align them...but it needs to be really poor... why include these frames?). 

    I show in my videos you can just blink your frames and in 3-5 minutes just get rid of the worst ones. THe ones with clouds...the ones that have doubled images...the ones that are simply blank. That's it. This is the simplest method. 

    I hope this helps give you a different perspective.
    Just to put a button on this... PSF Signal Weight (the default) is basically SubFrameSelector as a automatic weighting scheme. 
    -the Blockhead
  • Jim Smith
    Thanks for explaining. I will stick with Blink for now.

    Actually, now I think about it, I like the idea of seeing for myself what is wrong with the poorer frames rather than having them rejected by an algorithm that I don't really understand. I will have a clearer idea of what problems may be occurring during the capturing process.

    Jim
  • BlockHeadBlockHead
    Exactly... you always learn A LOT more by looking at the data.

    -the Blockhead
Sign In or Register to comment.