** WARNING - this image contains green colors
I always get the best tips on this forum and the latest to open up new processing avenues for me was learning about Starnet++, an open source star removal AI. I'm going to do the whole thing over and try my best to truly remove all evidence of stars, but as a first go I'm liking the results and will surely be revisiting a lot of old data. Here's what I did:
-Drizzle (I had to shooting at 2.8 arcsec/pix, so no MureDenoise)
-Stretch to .1 background (very light stretch)
-Run through Starnet++ (which, thankfully, was trained on a system somewhat close to the one use for this image (small refractor and CCD) so the algorithm worked very well)
-Use PS (for the first time!!) to clean the images as best as I could. I'm a total PS noob so I'm sure I can do better going forward. Each image took about 6 hours, probably)
-Back in PI, stretch to .2 background
-Combine with modified SHO palette
-Curves, Color Mask, HDRMT, LHE, MLT sharpen, TVGDN, ACDNR
-Convert to ICC profile
Please feel free to comment. If you don't look tooooo close, I think it looks pretty good for a first starless image.
Edited by drmikevt, 18 September 2019 - 08:20 AM.