Wow, this photo was a mess to process. I had OIII data, but didn't use it for two reasons. First, the data was ruined by bad guiding and second, there's just not enough OIII to justify using it. At least I didn't get enough OIII data to see any difference that it would make.

The next problem I had was that I accidentally shot my SII data in 2x2 bin (software bin with my camera). So my processing was a nightmare to say the least. I stacked the SII data, but couldn't use calibration frames as they were all 1x1. DBE was able to correct most of the gradient issues here. But the stars were all horribly bloated and oversaturated. So I drizzled the 2x2 data to make it the same size as my Ha data. Then I used Starnet to remove all the stars and I denoised the data as best I could.

With the Ha data, I drizzled it as well and processed two versions, one to be used for combination with the SII to make an RGB image, and the other was processed as a luminance image. I then downsampled them to work with the drizzled SII data. I also removed the stars from the Ha data with Starnet so I could use only the Ha stars in the final image, and get a good color. I made yellow colored stars by combining the Ha star layer in Pixelmath with one in red and another in green. Then I used the Curves Transformation to stretch the star layer and get the color where I wanted it.

The color combination I used with the Ha and SII data in Pixelmath was: Red - SII, Green - Ha*0.6 + SII*0.4, Blue - Ha. I tried other combinations, but I liked this one the best. I won't go into detail about all the other processes I used, but I had to use quite a bit of denoising on the RGB data because the SII was horrible. The Ha luminance looked really good, and was the only thing that saved this image fortunately, and the image looked much better after combining the luminance with the RGB data. After this, I combined the star layer back with the LRGB image using the Blend script in screen mode.

Finally, this image was cropped quite a bit to get the look I wanted. With the closer crop, the heavy denoising I did became apparent with that "plasticized" look in places. So I used the Noise Generator process to add noise. Then I upsampled the image with the Resample process by 200% with lanczos3 interpolation. I then added more noise with Noise Generator. This really helped even out the image and cover up the bad areas.

Overall, I think the image turned out pretty decent even though I screwed up royally with the SII data.