I don't often post incomplete works, but a lack of good weather in the forecast, and an eagerness to get some feedback and criticism has led me to go ahead and post this to get some thoughts. I am eager to improve and to learn.

I'm just starting to get the hang of narrowband imaging, and even at F2 it seems it really requires a lot of integration time. I started with shorter 2-minute subs, but have increased to 4 minute subs, and shooting at unity gain of 111 with an offset 8. I calibrated with an output pedestal of 150 which eliminates the black-pixel truncation during stacking.

I choose this target to be one of my first narrowband projects because of the unexpected amount of Ha I was able to see in my shot of [NGC 6871](https://www.astrobin.com/415835/?nc=user)

There ended up being far more Oiii in the region as well ... in particular the blue swoosh towards the center of the frame being the Wolf Rayet nebula 134... It represents the oxygen shell portion of the larger hydrogen shell that makes a ring from the nearby Wolf-Rayet star to the lower left of the blue shell. This ring shape is likely made by the stars stellar winds.

The Tulip itself on the right is such a beautiful feature, and I am working on acquiring data for it before season ends.

Current workflow:

* Process each stack separately using

* Dynamic Crop

* DBE

* Deconvolution (skipped on Siii as it has extremely low integration time)

* TGV Noise Reduction

* Initial stretch to nonlinear using Histogram Transformation

* Starnet to Remove Stars

Starless Processing of each stack:

* Histogram Transformation to bring the white point in without clipping

* Power of Inverted Pixels with lightness mask and very little smoothness

* HDR Multiscale to bring out the contrast on the brighter regions

* Stretched each channel independently

Combination:

* Straight up Red=Ha; Blue=Oiii; Green=Sii and adjusted curves of each until the combination produced a color balance pallet that felt evenly distributed between primarily the Ha and Oiii but some gently Sii (due to the very low integration)

* Created a synthetic L using Ha data and Oiii data

iif(

HaStarless>OiiiStarless,

HaStarless,

0.4*HaStarless+0.6*OiiiStarless

)

* LRGB combine L with the color data with chrominance noise reduction enabled

* Curves to brighten and adjust saturation to taste

* Light Local Histogram Equalization

* Recombined stars with Red=Ha; Blue=Oiii; Green=Sii

* Re-added stars with $ShoStarsless+0.85*ShoStarsOnly

* Light TGV Noise Reduction