Earlier this month I did an HOO narrowband version of this object. Although I had collected some SII data there wasn’t enough to use yet and I really liked the natural feel of the HOO image. I liked it well enough that I had an acrylic print done but that’s a story for another day.
In mid-August I was able to finish collecting enough SII data to finally process this. The first question was how to combine the data. I looked at all six combinations and liked three of them (SHO, SOH and HOS). I finally decided on the HOS palette because by putting the hydrogen alpha in red, it gave the image at least a slight grounding in reality. It also worked well with the story I wanted the image to tell which was destruction. The colors here really bring out the filaments and the wispy clouds of the gas. At least it does for me. Your mileage may vary.
I also rotated this 180 degrees from the earlier version. In the earlier version I felt the Veil was the main subject and the Triangle the supporting actor but in this version, I felt like the red and green wispiness was the star (sorry, pun not intended for once). Pickering’s Triangle drew the eye along into the wispy gas and the Veil provided an anchor to the image. I didn’t like this framing as much on the other image nor the other framing on this image. It surprised me how much the color choices influenced how I felt about the framing.
I also wanted to try the tone map processing technique. One of our club members had presented this a while back but I couldn’t conveniently try it because it depended on the starnet module in PixInsight and starnet would not run under macOS Catalina and I was never motivated enough to deal with exporting and importing images to multiple programs. But, with PixInsight 1.8.8-6, starnet now runs on the Mac inside of PixInsight! That seemed a perfect opportunity.
However, it crashed and burned. My final image was promising but the noise was too obvious and the background had become mottled. I tried a few more times but finally gave up. Apparently I need to learn more about noise reduction.
Instead, I decided to try an idea that I had been thinking about for a while: processing the stars and nebula separately. The idea was to split the image apart and do a stronger stretch on the nebula and weaker stretch on the stars. Then recombine them at the end. In theory that should keep the stars small and eliminate the need to shrink the stars later.
I put the technical details on astrobin but for a first attempt this didn’t go badly. There are two tradeoffs. The first is that any star that makes a halo that intersects with the nebula will cause artifacts. The nebula, with it’s stronger stretch, will have the halo, but the star with it’s weaker stretch will sit well inside the halo and look strange. To mitigate that I stretched 52-Cygnii in the veil itself a bit harder than the rest of the stars and it sort of works. It looks.a little artificial but it isn’t horrible.
The second artifact is that the stars essentially look like they are all in front of the nebula now. Any depth clues in the original image have been lost. I don’t think that is a problem in this image but others might be.
Finally, for this image I only used the stars from the hydrogen-alpha master. This give me pure white stars without any color. Since narrowband stars can be oddly colored I didn’t view this as a problem but some might not like it. Of course, you could have pulled the stars from a merged version. Alternatively, this might make replacing the stars with RGB stars easier since it’s relatively easy to pull the stars out like this. I may have to try that on a future project.