Data is from 2020-08—around 17 hours worth across four nights—and captured with a stock Sony Alpha A7R IV with the Radian Triad Ultra filter and the William Optics RedCat 51. Lots involved, here, which is less-than-optimal, or even problematic, to say the least. I was freeing up storage space by browsing large folders, noticed I had a mass of unarchived data from this session, and I couldn’t help but re-process it to see what might be possible with my current post-processing knowledge. I remembered being pretty happy with it—thinking that it exceeded my expectations—at the time.

My goals were to make the best of it, and to try out some various approaches to creating a tri-color “narrowband” image with channels derived from the color data, in the general spirit of how I usually edit SHO. Before I started using a mono sensor and filters, I was pretty intrigued by some of the approaches people were taking to create colorful images using these narrowband filters at the time.

This worked out better than I expected, and I’ve ended up with an image I am delighted with. I suspect the RTU, being capable of picking up Hβ, made a helpful contribution to being able to differentiate synthetic channels from the blue and green channels captured, and I also suspect having such a clean, low-noise sensor, with calibration frames and plenty of data, helped to make the most of signal vs noise concerns that one would expect to significantly hobble red channel data with such an aggressive narrowband filter (and maybe this would go quite poorly taking a similar approach on an older, noisy sensor). (Same sensor, here, as in the ASI6200MC, hobbled by the stock IR cut). The red channel was still the weakest channel—noticeably so—but I was able to bring up relatively faint structures and produce a reasonably clean result. I’m also delighted with the color balance I ended up with. A sort of “SHO”-derived color palette which pushes into “HOO” territory with depth and character in the warm to cool tone transitions that I like to aim for.

Basic Details
Sony A7R IV + RedCat 51
Filter: Radian Triad Ultra
iOptron CEM-40EC
ZWO ASIAir Pro + 30mm Guide + ASI120MM Mini
Salt Lake City, UT, USA (Bortle 7-9)
Dates: 2020-08-07 to 2020-08-10
17 Hours Integration Time

Editing
Detailed Processing Notes
https://tinyurl.com/JPNGC7000
Early Editing Example Screenshots
https://tinyurl.com/JPNGC7000eg

Processed in PixInsight and Adobe Photoshop. Four nights of data stacked separately. Integrations aligned and background extraction (ABE+DBE) on each separate file. Normalized with NSG and integrated with NSG-generated weights. SPCC, mild application of deconvolution via BlurX (due to data leaning toward undersampled), stars removed, NoiseX, GHS for stretching, some channel adjustments via GHS and Curves (contrast, color balance, and saturation), LHE and other detail enhancement (e.g. dark structures), adjustment to outer details which suffered a bit from flats over-correction (using a mask derived from the light gradient removal).

Lum and RGB channels were split. “Hα” created from red channel, “OIII” created from green and blue channels, and a third synthetic balancing channel created with a counter green-blue balance (possibly aided by the RTU picking up Hβ). Luminance edited for details, structures, noise, contrast. “SHO” created via PixelMath using synthetic channels with some trail-and-error adjusting the third synthetic channel to bring out the warm-cool transitions with more color depth, then colors adjusted to my preference of an “HOO-like” blend similar to what I would do with proper narrowband data. Blended with luminance, sent to Photoshop. There, additional cleanup, color balance, detail enhancement using adjustment layers and masks. Small fuzzies like IC 5076 were separately edited in Pix, brought in, then blended.

Stars (from SPCC) were slightly corrected with BlurX, correction, no reduction, halos slightly brought up (I want to feature them in the image as they add interest to the region left of frame). Soft stretch via GHS. Curves to slightly match target frame intensity. Separated via StarX, then added to a background nebula image from PS. A second copy of the nebula was used with Bill Blanshan’s PixelMath star reduction method for a little more reduction. Images of nebula with and without stars used in PS, with subtract method, to create a stars image, that blends beautifully with the Linear Dodge (Add) blending mode. Small adjustments to stars layer/folder with adjustment layers employing a clipping mask, and final adjustments to image.