Elongated and Saturated Stars in Your Images? Here's How to Fix Them in Siril
Вставка
- Опубліковано 30 вер 2024
- In this video, we will explore the common problem of elongated and saturated stars in astrophotography and learn how to fix them using Siril. Elongated and saturated stars can occur due to a variety of reasons such as overexposure, tracking errors, and atmospheric turbulence, and can severely impact the quality of your astrophotography.
I'll cover the new tools, "Desaturate Stars" and "Full Resyntehsis" in Siril to correct these two issues.
👽 Subscribe to my Channel ➔ www.youtube.co...
Make a channel donation or join one of the membership programs.
👽 UA-cam: / @deepspaceastro
☕ Buy me a Coffee: www.buymeacoff...
👕 Deep Space Astro Merch: deepspaceastro...
I make a small commission at no cost to you when you make a purchase through my affiliate links below. It's a great way to help support the channel!
🔭 High Point Scientific Affiliate Link: www.highpoints...
🛒 Amazon Affiliate Link: amzn.to/3YWWnwH
📚 Siril Tutorials
🗐 • Siril Tutorials
🏘️ Social Media:
🚪 Facebook ➔ / @deepspaceastro
🚪 Tiktok ➔ / deepspaceastro
#DeepSpaceAstro #astrophotography #siril
Please consider supporting this channel on Buy Me a Coffee!: www.buymeacoffee.com/deepspaceastro
Done.
@@yervantparnagian5999 Thanks so much!
Great series Rich, so much more information than other channels and will save me quite a few bucks by not having to purchase Pixinsight. Keep up the good work and I hope your channel grow heaps.
Thank so much! I hope it does too! ;-)
Hey Rich. Sitting at the bottom of Africa makes it near impossible to get the latest tech and equipment that needs hard currency. The more I am grateful for people like You and Cyril for investing your time and effort in projects like this. Internal thanks :bow: :bow:
Oh man that's gotta drive you crazy. Thanks for the kind words!
@@DeepSpaceAstro At least I will not change much 😉, stay well
Hi Rich. How are you?
Is there a way to make it leave certain stars alone? Sometimes it replaces a promient star but makes it look fake (or at least not as natural). I would like for it to leave certain stars alone. I tried deselecting the star, but it ended up removing the star completely when I did that (instead of leaving it alone, it removed it completely). Thanks!
Hmmm. I haven't tried to see if that was possible so I don't want to 100% say no. Sorry I don't have an answer for you.
Thanks for these detailed walk throughs of the Siril features, and your processing tips!
Glad you like them! Thanks!
Thank you!!!
Works great! Even handles some of the coma that was present with my coma corrector not in the right position. Thanks!
Glad it helped! Thanks!
Can i ask why osc and not a drizzle pre process?
For saturated stars? If that gets you the desired result, then you could do that. This video was just showing the functions of Siril to help with that. There's almost always more than one way to do things. 😉
@@DeepSpaceAstro i sort of meant in general, but now i realise you were doing it for demo purposes. Really appreciate your videos keep them coming please. Ill catch up eventually such a lot to take in :)
Brilliant thanks again for your excellent informative videos .
I really can't afford pix right now and I think my imaging doesn't yet warrant that expense .
Thanks!
i love my life but not my situation
Friend of mine gave the link to this video 📹... GREAT!! Tomorrow I'm going to reprocess my M92 result I made during my holiday! Thanks a lot for this video!! 👍🏻
Awesome! Hope you find it useful!
We need people like you, especially newbies like me. Much appreciated your work and contribution!!!! a huge shout out to you!
I appreciate that!
I’m having issues with color calibration all my info is right but won’t plate solve. Thinking it’s my starts
Try drawing a large selection around the center of your image, and then run PCC. You can also try the image plate solver under the hamburger menu, top-right of Siril. If it plate solves, then immediately try PCC after.
Thanks again for taking the time to make these videos. It definitely helps a (relatively) new person learn!
You're welcome! Thanks for watching!
So can we use this for coma aberrations in the corners? If so that converts a whole load of previously unsuitable f1.4 lenses into awesome milky way astro lenses. Im thinking things like the Nikon 24mm f1.4g i loved that lens for everything except astro
Honestly haven't tried it in that scenario, but I would expect it to work.
Blown away by this. I used in on my Cone Nebula and it worked perfect. Looks like the latest version of the recomposition has separate "Apply" buttons. So, I guess they fixed it from closing? If it still closes, correct me if I'm wrong, but I'm thinking of a work around. After the Starmask is created, I can bring each photo into GHS make them nonlinear so PS can "See" them. Save each as a 16 bit Tiff and bring them into PS. Once finished in there, bring them back and reconvert into 32 bit for Siril and THEN do a Full Resynthesis to fix star issues and finally do a Recombine? Guess I have something to do tomorrow. lol
Yep they fixed that issue a few months ago.
I'm a beginner in Siril! It's a perfect Video, thank you!
I've to reduce the speed of the video slightly and must with subtitle watch more often!
Best regards from lower Austria!
Glad it was helpful! Thanks!
Hi Rich. I'm just getting started with SIRIL and your videos are very helpful. I'm a little stuck at 9:30, where you remove the stars. The, "execute" button isn't lit up and can't be selected. Any ideas? Thanks. Tom, Tucson
That usually means you don't have Starnet downloaded and configured properly, or if you do, your processor may not have the AVX instruction set. If you're confident you have it setup correctly, watch this video to see if it's the issue with your CPU. ua-cam.com/video/RLvmcwYX2K0/v-deo.html
This is exactly the video I needed today, it was spot on! Thank you so much for your hard work!
You're so welcome! Thanks!
Thanks for the time and effort you put into this, I just started using siril thanks to your easy to follow instructions... try to do almost everything in photoshop before, but siril is way better. Thanks again
Glad to hear that! Thanks!
What about avoiding elongated stars during acquisition?
By the way, Blur XT does this automatically, even if elongated stars only in corners.
Sure, avoiding that or reshooting the image will always be the best solution, and I mentioned that briefly in the video. This is just a way to correct the stars if you can't reshoot, like if you went to a dark site hours away, and didn't realize the stars were elongated. You can fix them since chances are you wouldn't head back out to that site again to reshoot.
And yes BlurXterminator can do this, but that's only for PixInsight.
Hey Rich. Howzit? ONLY since recently, Siril became very picky with stacking frames. I had many images as the one you showed in the vid, however since I reloaded 1.2 beta it will not stack frames that have even only slightly elongated stars. So a question, how do you get an image with "badly elongated stars"?? I know, I should do manual stacking, but I only recently moved from piles of mussel shells to the abacus, and one of the upsides of Siril is it has automated script. (-;
BTW, do have your coffee with milk?
First, I'd recommend installing the latest version of Siril 1.2.0-RC1 although their website is currently down at the moment. Second, take a look at this tip from one of my previous videos and give that a try. ua-cam.com/video/BOVLBVvHTdo/v-deo.html. Oh and coffee is always black. ;-)
Wow, that star full resynthesis is unbelievable . Thanks for making these videos !
But I was wondering, if those elongated stars are round again, are they on the exact position in the image ?
Yes, absolutely
Brilliant stuff - just a quickie - how do you deselect a circled star selected Dynamic PSF eg. the bright core in NGC 7023 Iris Nebula ..
Thanks! To remove the star, click on it and then click the minus button in the bottom-right corner.
Another stellar video! Thank you!
Glad you enjoyed it! Thanks!
Well. I will have to test it more but this function removes elongated stars by replacing them with fitting round drawings of stars. That alone seems a bit questionable, but it also heavily depends on the correct star recognition. And it often recognizes doubles as one star, killing your data.
For me, this is not the way. I like pretty pictures but the stars should be correct.
True and I'm not saying people don't need to worry about their data. It's just a way to fix bad results if you can't reshoot for some reason.
Still not getting it on the star resynthesis.
Is this done on each sub or do I try to stack and then do it?
It's ran against your stack. You can see I selected the result.fit file here ua-cam.com/video/VNAcOcpubgQ/v-deo.html which is my stack.
Amazing!!! Thanks for sharing.
Thanks and you're welcome!
Great stuff, thanks, so it may not work with Newtonian difraction spikes then?
Thanks! That's good question. My first though was no, but maybe it wouldn't pick up on the diffraction spikes?
Great video and your instructions are easy to follow. However, I was trying to apply this to an image of M13 and my starless image of M13 has a blur automatically applied to what remains of M13 and when reconstituted with the star mask, the center of M13 is blurry and looks horrible. Do you know of a technique that can work on star clusters like M13, M4 and others?
Thank you! I wouldn't recommend using Starnet to remove the stars from a cluster, and that may be why you're seeing the results that you are. If your stars a elongated in a cluster, I don't have or know of a way to correct that.
Thanks for the reply Rich. I didn't explain myself very well. The shot is a wide field shot with a AT72 scope and a Canon 77D. I only cropped it just enough to remove the stacking artifacts, so there is a large star field surround M13 and my problem is with the stars way outside the cluster.
No console logs in 1.2.3 either.
There never were any logs written, just what's in the console screen. I think you can start Siril from a command line and have it write out the console screen. Is that what you're talking about?
Hi, Rich, I just found your videos and love them. I've been trying to learn Siril and the video have helped tremendously. I was especially happy finding the one about fixing elongated stars. I had great success with it until I tried it on globular clusters; can't get it to work right. It seems to treat the central condensation as a nebula. When recombined the central part of the globular is simply an over blown mess. I'm following your script to the letter but I'm probably doing something wrong Any suggestions?
I haven't tried it with clusters. It may just be they're too close together. Sorry, I know that's not much help.
Well, it's new territory.@@DeepSpaceAstro
Great Video! Thanks!
Glad you like it! Thanks!
Dam....great job...That simple short tutorial unloaded a lot of editing power I have on a backlog of data. Results are comparable to pixinsight. But cheaper.
Thanks man! Siril is capable of so much more than people know.
@@DeepSpaceAstro I am on version 1.2 beta. What is your workflow regarding the stretches? Do you do all your stretching in star processing for data that has some star issues? Or do you do your stretching then go to star processing?
Seems like stretching first would cause more issues but I am not sure.
Thank you ! Great ! Very cool !
You're welcome and thank you!
👍!
Very interesting video Rich... anyway, as always.
I'm thinking about how to use this feature in case of some shots, during the night sequence are bad (such as an example a gust of wind) do you think it'll be possible to recover these shots intead to reject all, do you think it'll possible and then join to the good ones?
One more thing regards to the color of the stars after the synthesis stars process, in your video seem that alla stars are white, I'm wrong?
Ciao
Roberto
If you're talking about bad individual exposures that are rejected during registration, you can try the setfindstar command. I show how to do that here: ua-cam.com/video/BOVLBVvHTdo/v-deo.html
Yes my stars are white but I beleive that is because I use my L-Enhance filter when shooting. I don't remember for sure since I shot those images last year.
@@DeepSpaceAstro thanks a lot my friend for your support 👍😉
You're welcome!
I followed these steps for the saturated stars (ie, not script, just menu) and with D-PSF I got one maroon star. I next ran Desaturate Stars but the maroon stayed even after clearing D-PSF tool, then re-selecting, getting maroon again, and even running Desaturate again. On a zoom, I can see the star changing shape on each iteration. Is that just a seriously sick star, keep iterating?
Can't say that I've ever seen that happen. It's just one star but it desaturated others and left you with this one?
@@DeepSpaceAstro actually on closer inspection, nothing is happening.
User error. I was going back to old folders and was starting with .tifs that were pre-stretched output for PS.
But if I may… would you Deconvolution first before the Desaturation if wanting to deconvolute (for benefit across the image)?
Is there a way not to lose my spikes? Because I use a Newtonian so with the full resythesis I lost my spikes😢
I don't believe so, sorry.
@@DeepSpaceAstro ok thanks for the answer 😊
Desaturate before Deconvolution, or other way around? Same question for elongated tool?
I don't know that it matters. Try it both ways and see what gives the best results.
@@DeepSpaceAstro I’m assuming Deconvolution first because the un-deconvoluted star becomes the template for deconvoluting the whole image including nebula contours. But that’s just my guess. Must test.
Oddest thing, I was (admittedly rushing) to process an NGC6888. Stacked image (r_pp_*stacked.fit) and when I ran Resynthesize all, the whole nebula disappeared. I've used Resynthesize before on images pre-Starnet removal (I think!)... did I do something stupid, skip a step or something?
Reason I was doing this was same logic as Deconvolution... reshape before Starnet to avoid artifacts around the stars if I were to reshape them after Starnet and then not have a good fit when the stars come back in.
Been trying to script this all day, do you think these stars look natural, they seem too perfect to me?
Yep, because they're synthetic. You can add some blur using the gauss command in the console. Something like gauss 1.2. Keep in mind there's no undo for commands issued at the prompt.
@@DeepSpaceAstro yes, tried that but still not great, really needs convolution but stronger than the deconvolution.
hi there i was just wondering if there is a way to apply this to the sequence? or do i need to fix each frame and resave the lights befor stacking... currently my stars are to obong where siril is rejecting all but 7 of 90 frames
I don't think so, but maybe this will help with your issue. This link will take you to the part of my tips/trick video that shows a possible solution. ua-cam.com/video/BOVLBVvHTdo/v-deo.html
thanks that did the trick needed to only drop it to .4 for it to work! learning alot from your channel keep it up @@DeepSpaceAstro
Awesome! Thanks!
Great video Rich! What do you recommend (if anything) for a star that remains saturated after running the desaturation tool? I've run into this a couple of times where it fixes all but one star. Would you just leave it and continue on, run desaturate again, or is there some other kind of setting I can change for it to fix them all the first time?
Thanks! If it's just one star I'd just leave it I think.
@@DeepSpaceAstro Perfect, thanks so much. Thats what I was thinking but I'm still pretty new to this. When do you work on your star processing? Is it one of the last things you do before you stretch or is it better to do it earlier in the process before noise reduction, etc.?
I usually crop, BGE, color calibration, remove green noise, then remove the stars.
WHAT??? This is awesome!!!
Yep! Cool stuff!
Really appreciate this and all your tutorials.
Glad you like them! Thanks!
@@DeepSpaceAstro started astrophotography about half a year ago using a tracker and modded dslr funnily enough I'm shooting the lagoon and trifid, going to just keep shooting it for awhile and see what I get. Though since I'm using a camera lens this has helped an insane amount because the stars can be a bit wonky.
Very nice. Tanks!
Thank you!
it seems starnet is down
Wow. Yeah looks like it is.
Fantastic videos -I've been watching all your tutorials. Thanks for producing.
Glad you like them! Thanks!
Absolutely the highest efficiency in explaining how such miraculous functions of SiriL can make one's jaw drop! Thank you so much for your video! The star elongation remedy is totally incredible. Does that mean when shooting emission nebulae, we can have longer exposure without worrying too much about star shapes? Previously, I would shoot 1-min subs thanks to my tracking accuracy. I'm wondering maybe I can try a 3-to-5-minute approach.
Thanks! It's not intended for that purpose, but it's up to you if that's the approach you want to take.
So glad I had Google recommend your Site ! I'm an old SiriL User and I've been a little wary of the new version . Now I can watch your Vids and learn a few things . Nicely done . Thanks a million . Subscribed of course . /SRK
You're welcome and thank you!
Truly amazing.... seeing was believing! Another great video.
Thank you!
I really enjoy your videos. Excellent recommendations that make me go right back to reprocess my images. Kudos!
Glad you like them! Thanks!
Great video! Just starting with SIRIL and this was really easy to follow. Thanks so much
Glad you found it helpful! Thanks!
Amazing!!! Thank you again for all your help understanding Siril and its many many features.
You are very welcome! Thank you!
Oh man, this might help me avoid trying to fix my tilt issues that I’ve been avoiding! 😂 Thanks for the video!
Give it try. Good luck! Thanks!
Good man, thank you for this tutorial its a savior. Much appreciated! Subbed
Glad to hear it helped you out! Thanks and welcome!
Watching this again to fix some IR bloat, dropping a comment. Thanks for the video!
Glad to help!
Great update Rich. This is definitely giving PI a run for its money. Great work and really useful for us all.👍
Thanks Perry!
Excellent series covering all the new features in the upcoming SIRIL.
Thank you!
Great informative videos Rich. This is all new to me but you are getting me up to speed far quicker than I thought possible. Thanks so much.
Great to hear! Thanks!
I am so excited for these new features.
I hate processing stars and these features look like they will solve a lot of my problems.
But I am having an issue with my scope with chromatic aberration on my stars. My images don't have the colours aligned correctly and it causes noticeable issues causing colour bleed on the stars where one side will be tinged blue and the other green or reddish dependant on the stars colour. I believe it's caused by a combination of the scope being an achromatic doublett and that I may not have my back spacing dialled in perfectly. I'm trying to resolve the second cause but its never going to be perfect due to the achromatic optics.
I shoot in broadband using a Baader moon and skyglow filter. I mainly chose it because I loved the way it kept natural star colour. But I can't really show it off because of the issues I'm having. Do you know if these new features added could fix this? Or is the only way to fix it post shooting going to be to stack the rgb separately to align them correctly?
Not sure how the resynthesis would handle it. They say the star colors won't be lost, but maybe the synthetic stars will have less of the tinged colors. Give it a shot. It runs relatively quickly. Then blend them back in with the recomposition tool. Maybe keep the star mask stretch light will help as well.
Another great Siril tutorial, thanks Rich! Do you know if it's possible to use this as part of a script so it can 'fix' elongated stars prior to registration? Registration won't include the light frames if the stars are too 'out of shape', unless there's a way around this? Clear skies!
EDIT: Try this to see if it gets the images registered. At the Siril command line (bottom right corner) type without the quotes "setfindstar -roundness=0.1" then run the OSC-Preprocessing script. If that gets things going, then open the result file and run the Fix Misshapen Stars script and see what happens. It may not be 100% though. Let me know.
I've been looking at this, and you have my wheels turning. lol I'm "kinda" close to a resolution I think, but no promises. In the mean time shoot me an email. Address is in the About tab on my page.
Another good video, thanks. I had messed with both features but didnt realise you had to do the dynamic psf. thanks for the clarification and demo.
Thanks and you're welcome!
Thanks a lot! I also have a few good/bad set of data where the DSO is fine but stars are elongated, can’t wait to try this and see how it works.
You're welcome. Let me know!
Great, problem solved. Thank you
You're welcome!
this channel has been the most helpful for this newbie, no doubt about it.
Glad to hear that! Thanks!
You're really advancing the art by providing all of these fantastic tutorials, well done! One question - maybe I missed it; is there a way to control the amount of shrinking the stars get with the resynthesis? I have found that my stars end up looking like perfect white round dots. Did I miss a step? I'd like them to be rounder, but still have the color and a bit more of a "natural" appearance. Thoughts? Thanks again!
Thank you! Use Generalised Hyperbolic Stretch with the Human or Even Weighted Luminance for the Color Stretch Model to help with the star color. You can try to add a bit of blur to the star mask by typing gauss 1.2 (play with that number too) at the Siril command line. Save first because there is no undo when using commands. You could also try my script to shrink the stars. ua-cam.com/video/Na6GzKozpCI/v-deo.html
@@DeepSpaceAstro Thanks Rich, I'll try that.
Thank you, excellent video as usual
Glad you enjoyed it! Thanks
Holy cow! You saved one of my images. This is so awesome!
It's a cool feature of Siril isn't it? Glad it helped you out!
Will it also fix the little star comets caused by coma?
Give it a try. I'm not sure.
Hey Rich,
I heard in one of your videos that you have a M50 MK2, right?
Does it support the ASIair plus? I also have a M50 2 and I was thinking about to buy an ASIair, but how the M50 hasn't the intervalometer port. I'm not sure if it can work integrated.
I don't think you can use the M50 with the ASIair for the reason you stated, but I could be wrong. You can use a Bluetooth intervalometer app to control the camera though. I just don't think there's a way to tie the two together. If you haven't seen it, I have a video for the intervalometer app and M50.
@@DeepSpaceAstro I was researching about this topic and I found out the ASIair controls the camera through the USB port. It contains a shutter port, but it is not necessary. I also saw in their website that it is compatible with the M6 MK 2. This camera has the same chip as M50 MK 2. I believe it can maybe work. 🤔
@@meianoitee01 Nice! Thanks for letting me know your findings!
Hi Rich thank you for the informative and inspirational content. If running both processes is there a more optimum order to run them?
That's a good question! I would say desaturate first, but it probably doesn't matter.
@@DeepSpaceAstro ok great thanks ill test it out.
Awesome Rich. I must give this a try. There are some options in Photoshop that are effective in doing this but this is incredible. As always thanks for sharing 🤝
Thank you and you're welcome!
@@DeepSpaceAstro I’ve just tried this out today on an image of M31 taken last year. It was ‘incredible’. I actually took the starless image and the star mask out of Siril having saved them as .tif files and imported them into Photoshop. The stars looked a little unnatural but when I used ‘less crunchy more fuzzy’ in the Astronomy Tools Action Set (Pro Digital Software) they were transformed 👍
Fantastic! Thanks for the tip!
Thank you for your videos 👏
You're welcome and thank you!
Do you think I can use my starlayer from StarXterminator to make them round?
Hmm. Maybe. It would have to be a 32bit FIT file for it to have a chance to work. Give it a try and let us know.
@@DeepSpaceAstro Tried it with a starlayer from StarXterminator and it works. First converted the tiff to a fits. I have mixed feelings about the result. Elongated stars are completely gone and are nice and round. But when I add in the nebula the final result looks ’fake’ to me. Maybe it needs some finetuning.
Yeah I'd agree, it's not perfect, but if you can't reshoot it's probably the only option. A couple of others here said they got good results in Photoshop using the "Less Crunchy More Fuzzy" action from Astronomy Tools. Give that a shot if you have it.
this is a game changer for me! great video
Thanks!