Preprocessing- "reject bad images" not working

Image processing, astrometry, photometry, etc.
Post Reply
Forum_2016
Posts: 219
Joined: 08 Dec 2018, 13:30

Preprocessing- "reject bad images" not working

Post by Forum_2016 »

Hello,

I am currently experimenting with "lucky imaging". As it is tedious to sort thousands of images by hand I would like use the "reject bad images" function in "preprocessing". To me it looks like this function does not work because even when an image clearly exceeds either the value for "star elongation" or "stars FWHM max." it is still used by preprocessing.
Do I have to do anything special apart from checking the boxes for "star elongation" and "stars FWHM max."? I have also tried this function with well exposed images to make sure that the program had a good chance to evaluate star shape and size, but even with these well exposed images AstroArt added all frames, regardless of the settings for both values.
Any hints, tips or tricks or does this function work at all?

Thanks
Stefan

Forum_2016
Posts: 219
Joined: 08 Dec 2018, 13:30

Re: Preprocessing- "reject bad images" not working

Post by Forum_2016 »

Hello, both options work in Preprocessing.
Just be careful about the parameters: for example, the default FWHM max of 4.5 is pretty big, suited for multimegapixel images. For a common CCD image a value of 3.0 is better: just open a typical image that you want to reject and measure the FWHM of a star.
Please note that the first image of the sequence is always taken, and used as reference.
About "Star elongation", the elongation is calculated as the X/Y ratio of the star width, so a star trail tilted 45 degrees will not be detected.
You may post here some jpeg images that you would like to reject. To find the problem, I suggest to use only one rejection method at a time.

Forum_2016
Posts: 219
Joined: 08 Dec 2018, 13:30

Re: Preprocessing- "reject bad images" not working

Post by Forum_2016 »

Hello Fabio,

thanks for the quick answer.
I have tried various settings, but even at the lowest possible settings for elongation or FWHM images are not rejected.
I have put a part of an imaging series here
http://ccd-astronomy.de/M27.zip
where the autoguider seems to have lost the guidestar at some point.
Still none of the images is rejected even if I choose the lowest setting for elongation or FWHM.
This series was taken with a colour camera, but the same happens (or better doesn't happen) with a mono camera.
Even if I have bloated stars at 2m focal length with small pixels and bad seeing no pictures are rejected even with FWHM set to 1.5.

Maybe there is a problem with the choice of stars? How does AstroArt pick the stars it wants to use for those measurements?

Thanks,
Stefan

Forum_2016
Posts: 219
Joined: 08 Dec 2018, 13:30

Re: Preprocessing- "reject bad images" not working

Post by Forum_2016 »

Hello Stefan, thank you for the interesting sequence.
The correct rejection method would be "Star elongation" but in this sequence it fails because the trails are wrongly recognized as double stars, see below. This is a limit of the algorithm, it works only on "single" stars elongated.
So, you must use the option "Confirm each image" in Preprocessing to reject manually the bad images, this will allow to also reject the satellite trail.
Press SPACE or ENTER to accept an image, or ESC to reject. You can do it very quickly.

As a test, you can see the auto rejection at work if you enable the option "Binning 2x2" in Preprocessing/Demosaic, because that binning will merge enough the trailed stars.

My best,
Fabio.

einari
Posts: 44
Joined: 28 Jan 2019, 10:46

Re: Preprocessing- "reject bad images" not working

Post by einari » 28 Jan 2019, 10:49

Coming back to this older thread.
I stacked frames with AA7 using reject images and it seemed to accept all even if I set FWHM threshold as low as 2.

einari
Posts: 44
Joined: 28 Jan 2019, 10:46

Re: Preprocessing- "reject bad images" not working

Post by einari » 30 Jan 2019, 07:03

Update: I lowered the threshold to 1.5 and only one in stack, then 1.8 and 3 were rejected from stack.

Dunmunro
Posts: 225
Joined: 10 Jan 2019, 01:11

Re: Preprocessing- "reject bad images" not working

Post by Dunmunro » 30 Mar 2019, 01:06

I was experimenting on a 30 image stack and I found that the value for FWHM is very sensitive while elongation is insensitive but FWHM rejection required elongation of 1.1 or greater.


Elogation = 1.1
A FWHM value of 2.086 would reject 6 images
A FWHM value of 2.085 would reject 7 images
A FWHM value of 2.081 would reject 8 images
A FWHM value of 2.08 would reject 9 images


unchecking Elongation and FWHM but checking Planet sharpness and entering a value of 80 resulted in 10 images being rejected, but with only a poor correlation with the FWHM setting, above.

Post Reply