Remove Scattered Light in PixInsight

[@alfogator]

There's a square in Florence where they used to burn people that performed magic like this! Luckily we live in better times and can appreciate the brilliance of Adam.

[@AdamBlock]

Best response! lol My brain did not compute for a moment after reading the first sentence. Thanks!

[@lee-van-cleef]

I used this to remove halos caused by an SII filter - you saved me $$$ Adam!

[@AdamBlock]

Well... it depends... if you become a member of my site... maybe not... but you will found more solutions and ideas. :) AdamBlockStudios.com

[@teninnykanal]

Adam you saved my Dark Shark Nebula! I had the same issue and solved my problem with your help. Thank you!

[@AdamBlock]

Ooooh WONDERFUL! I like it when I save a Nebula!! LOL I have done something for you that literally no one else could be solving this kind of issue (no KZbinr and no other "competitor" of mine). Is this not enough for you to consider becoming a member of AdamBlockStudios.com ?

[@teninnykanal]

@@AdamBlock Definitely yes!

[@Aerostar509]

For those who don't know how to blink 2 images, lay them over each other and the press CTRL + PAGE DOWN.

[@glentyan2505]

Well done Merlin, your channel is a great source of enlightenment and you are the great sorcerer.

[@starsips]

Great video and example! Thank you Adam

[@terryrobison]

Very nice Adam. I can get interesting reflections at times. I will give this a go. Cheers.

[@angelofranckgambino2458]

It is an outstanding and genial solution. Thanks for sharing it! And what if you have the halo in all your L, R, G and B channels more or less in the same positions ? Thank you in advance! 😃

[@AdamBlock]

Unfortunately then a different solution will be necessary. This relies on the fact you can isolate the issue.

[@prabhuvpop]

Hi adam, great video, I shot some 20 hours of witchhead nebula and I got reflections on all my LRGB filters, i'm wondering how to choose a reference frame to get rid of the reflections

[@Makrem]

Genius as always

[@albertmennone5646]

Great tutorial Adam, I occasionally get a flare from the edge of the sensor of my OSC which extends a good ways into the image. If I split the channels it is there in all 3, although a bit less so in the red channel, can this method be adapted to work in this situation?

[@davidserquera3331]

Hello Adam, I think there is a script called substitute with preview that could do the trick, is that right?

[@AdamBlock]

Substitute by preview Yes..but it creates new images and is multi-stepped. Give it a try- report back whose method is better. :) My expression writes directly to an image.

[@nicolast3499]

Thank you very much Adam for this amazing technique but I have a question concerning your process. To sumarize (if I understand correctly) you’re trying to replace some part of the green image with the luminance one. It seems to me you can achieve that with a simpler expression and without using preview : Just do a mask over the defective parts of the green image and apply it to the green image In pixel math create an expression with only the name of the luminance image Apply the pixel math to the green image Am I missing something?

[@AdamBlock]

The preview is to calculate a local area for Linear Fit. This yields better results than calculating for the entire image. Once you have a small area to work with as I showed, my expression allows you to "paste" a piece of an image in place. The mask suggestion works if you are operating on the entire image.

[@nicolast3499]

@@AdamBlock Ahh ok, I was missing that part. I get it now, thanks!

[@AdamBlock]

@@nicolast3499 And... on top of that.. .you will note I *did* use a mask to paste the bit of the image in order to make absolutely certain no edges would show up.

[@frankpaulic7811]

This works well and got rid of artifact on my Luminous image, but unfortunately the area I replaced has a different noise stucture than rest of image (clearly can be seen), since I referenced to a Blue binned 2x2 image. Both inside of box and outside of box has the same noise level in the Readout Mode, thus the Linear Fit worked well. It should be noted that I did not use the GAME mask, since everything freezes up when I open the GAME script. Possible I have an older version of PI. If I did use the script, I still would see different noise structure within image.

[@PopeAstro]

would this work for my mosaic that has light leak in the bottom right corner?? Its from the ZWO filter drawer i just got. it sucks because i spent multiple nights on IC443. Its very obvious in my flats, i have all intentions buy the membership very soon!!

[@AdamBlock]

Yes, it might if you have other successful images that *do not* have the glow. If the glow is in every frame of EVERY filter... then you are pretty much stuck in terms of a nice mathematical way to handle it.

[@PopeAstro]

@@AdamBlock shit

[@PopeAstro]

@@AdamBlock any advice?

[@AdamBlock]

@@PopeAstro Without seeing the data..I can't say. But based on your response..I would predict this issue is an acquisition issue...that is what you probably need to tackle and not worry about fixing a mess by processing. Image processing is GIGO through and through.

[@mattagee319]

Would this process work with a OSC camera?

[@AdamBlock]

Yes, but it is unlikely that whatever phenomena caused this specific issue would happen as cleanly on an OSC (color filter array).

[@PaoloBanci]

Florence is florence! :-D