100 60s frames with Canon 800D iso 800, Samyang 16mm f2 at f2.8, tracked with Omegon Minitrack Lx2, processed as following

***** Integration

lightvortexastronomy tutorial (https://www.lightvortexastronomy.com/tutorial-pre-processing-calibrating-and-stacking-images-in-pixinsight.html)

* CC defect list + master dark (sigma = 10)

* blink for meteors:1482, 1484, 1533,

* weighing: (15*(1-(FWHM-FWHMMin)/(FWHMMax-FWHMMin)) + 15*(1-(Eccentricity-EccentricityMin)/(EccentricityMax-EccentricityMin)) + 40*(SNRWeight-SNRWeightMin)/(SNRWeightMax-SNRWeightMin))+30

* img 1520 ref

* drizzle integration of last 50 images (to keep the stars from going wobbly)

***** Adding meteors

* pixelmath max(frames with meteors), resize 2x to fit integration, denoise with EZDenoise script

* create meteor_mask:

iif(

(d2seg(3567, 1088, 5636, 4742 ) < 4) ||

(d2seg(1204, 2819, 2878, 3849) < 10) ||

(d2seg(812, 4682, 2332, 4562) < 4),

1, 0)

MLT blur first 2 layers

* meld, per RGB channel: (meteors - 0.006) * meteors_mask + (1 - meteors_mask) * drizzle_integration

***** Linear processing

*** Initial calibrations

* Crop (just edges)

* linear fit color calibration

* DBE tolerance 3, avoiding the milky way

* photometric color calibration centered on Sadr,FL 16, pixel size 1.9, background neutralization 0.077

*** Denoise

EZDenoise script, lowering TGV exponent by 1

***** Nonlinear processing

*** Initial stretch

** mask creation

* stretch luminance w/ ezstretch, use starnet to extract a star_mask, histogram stretch mids 0.1, blur 2 layers of MLT

* new meteor_mask:

iif(

(d2seg(3519, 1050, 5589, 4704) < 4) ||

(d2seg(1147, 2774, 2839, 3815) < 10) ||

(d2seg(761, 4645,2290, 4523) < 4),

1, 0)

** chroma stretch

* duplicate image, mask with star_mask, apply correct magenta stars

* arcsinh stretch black point 0.074 stretch factor 174 //protect highlights

* repaired HSV script

* with meteor mask on, acdnr on default settings

** luma stretch

* pixelmath: (1 - meteor_mask) * luma + meteor_mask * (luma - 0.0008 )

* STF background 0.15 shadows 2 + histogram stretch

* extract luma, apply meteor mask, MLT denoise thresholds 4 2 1 0.5

** combination

* lrgb of the two

*** Starless split

* extract starless image (some artifacts, parts of the meteor trail, but light enough to deal with)

* clone stamp out larger star artefacts

* summer - starless = stars

**** Starless processing

***MLT stretch

http://www.stelleelettroniche.it/en/2014/09/astrophoto/m42-ngc1975-multilayer-processing-first-part/

**Initial (fine details)

* extract luminance out of starless, shadows 0.05 mids 0.35, s-shaped lum curve

* mlt - keep the first 4 layer

* left = starless - mlt

* with the luminance mask on left, saturation and RGBT boost

* saturate reds and purples with color saturation

* starles = starless + mlt + left, normalized

* apply the redpurple saturator again

**Second (large scale colors)

* new multiscale linear transform, keep 6 layers

* diff from original - left

* apply the redpurple sat again

* extract luminance from blurred image, to use as a mask.

* masked blurred image with the luminance mask, slight boost in RGB/K and decrease sat curves

* remove mask, desaturate yellow-oranges, saturate reds pinks

* pixelmath sum the 3, rescaled, back to original image

*** Starless finishers

* LHE kernel 512 contrast limit 1.5 blended 5 * starless + lhe normalized

* LHE kernel 256 contrast limit 1.5 blended 5 * starless + lhe normalized

* LHE kernel 32 contrast limit 1.5 blended 3 * starless + lhe normalized

* Dark structure enhancer, default settings

**** Stars processing

*** Curves dial down luminance

*** Star reduction

* extract luminance, use starnet to create a star_mask. Clone stamp out the bigger, constellation-defining stars. Save as star_mask2

* min(star_mask2, (1 - meteor_mask) = just_stars

* use decon functions to stf stretch and boost just_stars

* apply on stars, morphological transform 4 iterations of erosion, intensity 0.2

**** Finishers

* Pixelmath sum stars + starless

* with meteor mask on, raise green saturation

* MMT sharpen, 6 layers biases 0.05 0.05 0.025 0.025 0.012 0.005 on luminance