1 stack of 58 60s images, Canon 800D at ISO 800, Canon 400mm f5.6 lens at f6.3, iOptron Skyguider Pro tracker. 120 darks, 120 biases. Processed in PixInsight as below


Meteors: 
7154 (small, one end)
7173 (huge stripe, andromedaish direction)
7177 (3)
8085 (clear fizzle out)
8456 (small stripe)

* CC defect list + master dark (sigma = 8)

* Subframe selection
30*(1-(FWHM-FWHMMin)/(FWHMMax-FWHMMin))
+ 30*(1-(Eccentricity-EccentricityMin)/(EccentricityMax-EccentricityMin))
+ 20*(SNRWeight-SNRWeightMin)/(SNRWeightMax-SNRWeightMin)
+ 20
Eccentricity < 0.7 && SNRWeight > 1 && FWHM < 5. Keep frames 7173 and 7177 even though they failed the rejection conditions, for meteors

* star alignment: 
img 8411 ref, frame adaptation (let's see if this works)

* ESD integration, range exclude 
* drizzle integration, gaussian kernel\


*****Linear processing

*** Initial 
* Crop
* DBE tolerance 3

*** CC
* PCC on Pleiades, default settings pixel size 2, background neutralization upper limit 0.0012

*** Decon
* star mask creation: extract luminance get starmask with starnet, then range mask 0.1-1 (1). from stretched luminance range mask 0.55-1 (2). Pixel math max the 2 as binary_star_mask. 4x dilate, 6x convolve
* PSF - ezdecon script
* background mask  -ezdecon script and curves
* with background mask on, inverted, apply decon 100 iterations on luminance, deringing global dark 0.01 light 0.002 and local deringing with binary star mask, wavelet regularization

*** Denoise
Using <a href="https://jonrista.com/the-astrophotographers-guide/pixinsights/effective-noise-reduction-part-2/&quot; rel="noreferrer nofollow">jonrista.com/the-astrophotographers-guide/pixinsights/eff...</a&gt; as implemented by EZSuite. 
* EZDenoise, TGV 1500 edge 1e-5, default MMT


***** Nonlinear

*** Initial stretch
* Masked stretch, default settings
* using a contrasty luminance as mask, curves stretch


*** Stretched color mods
* with binary_star_mask protecting stars, ACDNR on chrominance, lightness mask, stdev 4 iterations 6
* CC background ref upper limit 0.125
* invert, SNCR green 1, invert again (to fix magenta hues in whites)
* SNCR green 0.8


*** Star mods
* extract star mask with starnet, blur 3 layers of a trous, curve boost
* with it on, curve saturation upwards. Clone stamp out big stars - small_star_mask
* using that as a start, EZStarReduce using morphological transformation
* with small_star_mask on, unsharp mask 0.9, w/ deringing

***MLT stretch
<a href="http://www.stelleelettroniche.it/en/2014/09/astrophoto/m42-ngc1975-multilayer-processing-first-part/&quot; rel="noreferrer nofollow">www.stelleelettroniche.it/en/2014/09/astrophoto/m42-ngc19...</a&gt;

**Initial (fine details)
* created a new multiscale linear transform, kept 5 layers
* diffed from original image to create a "blurred" version of original image
* extracted luminance from original, s-shaped RGB curve used as mask on blurred version
* used curves to create s shape in RGB (asymmetrical, like a camera curve) and pump up saturation a lot
* pixelmath sum the 3, rescaled, back to original image

**Second (nebula)
* created a new multiscale linear transform, kept 7 layers, and diff from original
* extract luminance from diff, s-shaped RGB curve. Use as mask on blurred version
* slight boosts in sat and RGB
* pixelmath sum the 3, rescaled, back to original image


*** Finishers
* EzDenoise, TGV 1500 edge 8e-3, default MMT
* LHE kernel 512 contrast 1.5,slight boost in sat, 50-50 mix with original
* LHE kernel 256 contrast 1.5, 20-80 mix with original
* dark structure enhance
* MMT sharpen, 6 layers biases 0.05 0.05 0.025, 0.025, 0.012, 0.005 with luminosity mask s-curved for more contrast