IMPORTANT UPDATE! I got a repository link working! Please add raw.githubusercontent.com/setiastro/pixinsight-updates/main/ to your repository in pixinsight to add it
Script and repository access working great. I've used it to substantially improve the color of my NB stars. I'm learning a bunch from this channel. Thanks for sharing.
More color with your script than Foraxx. Very nice! :) I didn't have a chance to run on linear data but notice on pre-stretched stars there's some red artifacts around some stars and other areas that may be leftover nebula remnants from starX.
Great to here. For me it ran quite a bit faster too. StarX does leav in some items. That is why I like using NB stars to make sure those extra NB bits get put back in my image. It can make a difference on some small strictures for sure.
Adding script now. Nice work! A quick note- your website has a typo for the Star Stretch script: "Star Stretch will take your non-linear stars only image and stretch it up to a non-linear state" should say "take your LINEAR stars"? Cheers
Great idea! I've run my test with linear and non-linear data but I can see that if there is dominant channel then those stars take that color. I have multiple only-red stars. Even the Orion's trapezium is only red.
Interesting. The stars thrmselves shouldnt be excessively dominant in one channel over another in their linear form unless there is a big difference in exposure lengths between filters or if the filters have different bandpasses. At least I havent ran into that particular problem yet.
@@setiv2 All my data are 600s , SHO 3nm Anltia filters. I'll email you some screenshots for the record. In the meantime I'm trying to understand how the script works (looking into the code) to see if I could find anything ringing a bell.
That will be helpful for sure! Nothing weird in the code. Some pixel math for a natural color star combination, scnr, saturation, then the stretch is a logarithmic stretch mathematically equivalent to midtoned adjustment in histogramtransformation.
Nice one! Will this work with one colour shot images, say for example for DUO NB filters like L-Xtreme? I guess will also have to perform SPCC prior to running this? Thanks!
Yes! There was someone on thebPixInsight Tips and Tricks facrbook page usong that filter giving it some comparidons with great results. You have to seperate your channels first. Here is thr link to look at his comparisons. Check the comments for his comparison shots with OSC facebook.com/share/p/Sj6ZVrnGpkdHNkNY/?mibextid=oFDknk
Always great to see people making processing tools, but I'm not sure why you'd need a whole script for something that can be done with a simple pixelmath expression 🤔
Thanks for the useful script. I tried the script on my narrow band stars. It gave reasonably good results with Ha + O3 but strange red colored stars when S2 was added. Any ideas? I am using ZWO 1600 MM Pro camera with Antlia 5nm filters for Ha & O3 and ZWO 7nm filter (supposed to be second generation "better than the previous one" for S2.
Yes. Your SII filter is letting in a lot more light. Did you linearfit the sii master to your others first? Honestly, with having a different bandpass on your SII you may want to just stick to Ha and OIII for the script.
Hi - I tried this process on an image using 2 x dual band filters. I achieved the 3 channels using DB Extract before doing Star X and then ‘NB to RGB Stars’. Unfortunately the results were not very good, with or without Sii, with very uneven star colouration. I tried different strengths with the sliders. Maybe this is something that is not possible to do or there is a better way to split the channels? I can get acceptable results using just the Ha/Oiii image, enabling your dual band option, so should I just stick with that?
Thanks for your reply. I tried your suggestion with similar results. To clarify - starting point was 2 master lights (Ha/Oiii and Sii/Oiii). Step 1. StarX - both images. Step 2. DB Extract on Stars only images. Step 3. NB to RGB stars (only using Ha and Oiii stars). Pulled back the stretch sliders from default. Outcome is mixed colour stars. I consider myself to be very much a learner in these matters, so it is likely that I am missing something obvious😢. Thanks for your patience.
Just installed the script. Will try using it in the next PI processing run. May it be possible that in the future the script can be installed in the usual PI "from repository" way so that in case that the PC gets replaced we do not need to download it again?
Would it be possible to support stars from OSC camera with dual band filter? How would you do this split RGB and combine green and blue into OIII then run through this script?
Yes. Extract your R G and B and then either combine the G and B or just use the G (probably has the higher SNR) and run it through the script with your Red for Ha and the G for OIII
You need to update pixinsight to the latest build. (1605 came out in march) you can get it from the pixinsight download site. That will fix that issue. Otherwise you need to delete the .xsgn file Hope that helps Frank
