I took what I thought was pretty good LRGBHSO data for the Eagle Nebula. The LRGB combination looks OK,
but I completely struggled with adding in the narrowband data. No matter what I did, I didn’t seem to be able to add in the rich detail from the Hα without losing the rich color from the LRGB. Here's the Hα:
The Hubble palette attempt turned out OK as well, but in retrospect, I think the Hα is probably stretched a bit too hard so I may go back and try a redo on the SHO version. It is clear there is a lot of detail in the Hα.
I’ve tried a lot of different variations for processing this year’s narrowband data for M16. The final colors are very sensitive to what I do in early steps of the processing. Obviously the amount of stretching applied to the various channels matters a lot. I have a version I like a lot that is mostly in golds, but ultimately I think the best rendition is this one where I try to make the dark end of the histogram similar for all three color channels and the OIII still shows up as blue. This is a really nice target for SHO because there is significant emission at each of the three wavelengths. RGB stars were extracted from my RGB image, shrunk a bit with Carboni’s “make stars smaller”, then added using screen mode in Photoshop. Several rounds of “less crunchy more fuzzy” were applied, with the effects masked out for the Pillars and other places where there is detail I didn’t want to lose.
But the point I really want to make is best illustrated by my various attempts to blend the narrowband data into the LRGB. For example, in this image I tried to use both Hα and OIII blended with the LRGB. I had thought that OIII goes best with B, but I think a stronger argument could be made for blending OIII with G. Be that as it may, this just didn’t turn out as well as I wanted although it was better than what I had gotten to that point.
Disclaimer: this image uses the same data as my last two posted images (in fact it combines the data from them). This is an attempt to get a “truer” color version of M16 but including the contrast and depth you get with narrowband filters. I used a combination of Hα and L for the L layer, Hα and R for the R layer, G only for the G layer, and B+OIII for the B layer. Each of the 4 layers was processed separately and each was stretched roughly the same amount. Combination was done using StarTools’ LRGB module. I did some slight tweaking in Photoshop using Levels, Hue and Saturation, and Vibrance but otherwise didn’t process much after combining the layers. This is not stretched nearly as hard as the SHO image was, and I think this version turned out better than my straight-up SHO. The gentler approach wins.
Then, while working on another image (M17, the Omega (or Swan or Lobster) Nebula), I remembered the Starizona web site (https://starizona.com/tutorial/using-an-h-alpha-image-as-a-luminance-channel/) has a tutorial on doing HαRRGB combinations and decided to review it. Their approach involves mixing (mostly) Hα with R and using that combination for the R layer and for L (method 1) or using the Hα by itself for L (but at ~75% opacity) with the HαRGB contributing the color (method 2). With my data, these approaches gave results I think are far superior to what I got in my feeble attempts to figure out the mixing:
I don’t usually repost the same data under a new heading, but after seeing how the Starizona HαRRGB methods worked on M17 I had to go back and try them on M16. These data look (in my opinion) dramatically better when processed using the Starizona methods so I’m going to break my own rule. I used my RGB image posted previously (https://astrob.in/352634/B/), added Hα to the R, and used either Hα+R (method 1) or Hα alone (method 2) for luminance. I tweaked the curves as suggested on the Starizona site, boosted the saturation a little, and used Carboni’s layer-masked “less crunchy more fuzzy” to make the stars less harsh. I think these look a lot better than what I posted earlier. This time I think I prefer the method 2 image for the detail it reveals in the nebulosity.
Date: 15-20 Jun 2018
Subject: M16, Eagle or Star Queen Nebula
Scope: AT8IN+High Point Scientific Coma Corrector
Filters: ZWO 31 mm diameter unmounted L, R, G, B, Hα
Mount: EQ-6 (EQMOD 2.000j)+PEC
Guiding: Orion Thin Off-axis Guider + DSI IIc +PHD 2.6.5 (Win 10 ASCOM)
Camera: ASI1600MM-Cool, -20 °C, Gain 139 Offset 21
Acquisition: Sequence Generator Pro 3.0.2.91
Exposure: 51x180 L, 20x180 R, 20x180 G, 20x180 B, 48x300 Hα
Stacking: Deep Sky Stacker 4.1.1 (64-bit) dark+flat+bias, κ-σ stacking with κ = 1.5.
Processing: StarTools 1.4.332: Combined R, G, and B in StarTools. Binned 2x2, cropped, wiped, developed, HDR, color, deconvolution. Stopped tracking to smooth. Binned L 2x2, cropped, no wipe, and developed followed by HDR. Deconvolution followed by untrack smoothing. Combined L with RGB in Photoshop. Used some layer-masked saturation adjustments, both in the master and in the red channel, to try and retain the red color. A little of Carboni’s Astronomy Tools’ increase star color followed by less crunchy more fuzzy. Green cast in stars removed using StarTools’ “Cap Green to Yellow”. Also some different choices in Curves for the luminance and RGB layers to better bring out the reds in the emission nebula. Added Hα as described above.
