M13 - Hercules Globular Cluster
I have returned to the Great Globular Cluster in Hercules. On the very few nights here lately I have been looking for a small quick target to shoot around the 5-6am time frame just before the sun comes up, (and while the bright full moon is high in the sky), and well M13 goes right close to zenith then, and since I love this glob it was an easy choice. Since the
first time visiting this cluster back in 2018 I feel I have learned quite a bit, and improved some optical issues with my OTA, but imaging at the smaller image scale of 0.5"/px required me to pull out some old tricks I had learned the first time. Please enjoy!
Constructive Criticism is highly encouraged as I hope to continue to improve my processing abilities.
Equipment:
[li]
Orion 8in F4.9 1000mm Newtonian Reflector
[li]Flocked / Primary mirror replaced due to turned-down-edge[/li]
[/li]
[li]
Skywatcher F4 Aplanatic Coma Corrector
[/li]
[li]
Astrodon I Series L, R, G, B broadband filters
[/li]
[li]
ASI183mm Pro cooled to -15C
[/li]
[li]
Celestron CGEM Mount
[li]
Self tuned / hacks to get guiding stable include:
[li]
Intentional offset polar alignment so dec always pulses in one direction
[/li]
[li]
Balance "west" heavy (rather than the recommended east) so that the ota "falls" onto the gear teeth rather than get "lifted"
[/li]
[li]
Factor Reset hand-controller daily (to prevent cgem from being possessed and forgetting where the meridian is on subsequent night)
[/li]
[li]
Dither in RA only
[/li]
[/li]
[/li]
Acquisition Details:
[li]
Images captured using N.I.N.A
[/li]
[li]
Red: 45x60s (+4 rejected)
[/li]
[li]
Green: 38x60s (+11 rejected)
[/li]
[li]
Blue: 40x60s (+9 rejected)
[/li]
[li]
Lum: 39x30s (+11 rejected)
[/li]
Total Integration Time: 2 hours 23 minutes
Imaging Dates: Data Captured over three Nights in early June
2020-03-08|2020-03-09|2020-03-12
The resulting image is a combination of the following steps:
[li]
Inspected all 197 subs for bad images with Blink after each session, discarding those with clouds, or trailing stars from poor stars.
[/li]
[li]
Calibrated all subs with their corresponding master flat and master dark
[/li]
[li]
Used subframe selector to weight all subs of like filter together using the following formula:
[/li]
(15*(1-(FWHM-FWHMMin)/(FWHMMax-FWHMMin))
+ 25*(1-(Eccentricity-EccentricityMin)/(EccentricityMax-EccentricityMin))
+ 15*(SNRWeight-SNRWeightMin)/(SNRWeightMax-SNRWeightMin)
+ 20*(1-(Median-MedianMin)/(MedianMax-MedianMin))
+ 10*(Stars-StarsMin)/(StarsMax-StarsMin))
+ 20 // (20 for RGB and 30 for L)[li]
Selected the best sub from subframe and blink to use as a reference frame in stacking
[li]M13LLIGHT2020-03-0806-27-20001530.00s-15.000.42_c.xisf[/li]
[/li]
[li]
Created a super luminance by integrating all remaining 162 frames together
[/li]
[li]
Integrated all Red frames together, Blue frames together and Green frames together to create masters for each color
[/li]
[li]
Cropped the stacking edges of the integrated masters
[/li]
[li]
Combined the Red, Green and Blue masters to create a color RGB image
[/li]
Superluminance Processing:
[li]
Dynamic Background Extraction
[li]30 Manually placed DBE points of radius 50 around the perimeter of the frame and around the cluster, avoiding star halos.[/li]
[/li]
[li]
Motion blur deconvolution to negate the effects of poor guiding at this high imaging scale
[/li]
[li]
Two rounds of Noise Reduction was done using TGV Denoise with a low contrast mask and an autostretched local support (First round with very strong edge protection, second round with very weak edge protection but very small strength)
[/li]
[li]
Masked Stretch was used with no clipping and a background reference created from 4 preview windows and the Preview Aggregator script. Histogram Transformation was used to bring the tail of the blacks in without clipping and perform additional lifting of the cluster..
[/li]
[li]
HDR Multiscale Transformation was used with 9 layers using the Median transformation and Lightness Mask to lightly give lift to the stars of the cluster
[/li]
RGB Processing:
[li]
Dynamic Background Extraction
[li]30 Manually placed DBE points of radius 50 around the perimeter of the frame and around the cluster, avoiding star halos.[/li]
[/li]
[li]
Background Neutralization using 4 preview windows and the Preview Aggregator script as the background reference
[/li]
[li]
Color Calibration using Photometric Color Calibration using the Average Spiral Galaxy as the white reference
[/li]
[li]
Two rounds of Noise Reduction was done using TGV Denoise with a low contrast mask and an autostretched local support (First round with very strong edge protection, second round with very weak edge protection but very small strength)
[/li]
[li]
Noise Reduction using MMT with a very protective luminance mask targetting luminance and chrominance
[/li]
[li]
Repair the saturated star cores with the HSV Repaired Separation script
[/li]
[li]
Arcsinh Stretch and Histogram Stretch to bring to non-linear
[/li]
[li]
Additional curves transformation to increase overall color saturation before final histogram stretch
[/li]
Final Steps:
[li]
The super luminance and the RGB image were then combined using the Channel Combination tool in CIE L*A*B mode to apply the Super Luminance to the RGB image
[/li]
[li]
Curves Transformation to increase contrast and colors throughout the field
[/li]
[li]
Manually removed a chrominance blimish in the bright blue star on the right using clone stamp
[/li]
[li]
Exponential Transform using PIP to slightly lift the background using a luminance mask to protect the highlights
[/li]
[li]
SCNR Green to remove some green color fringing on some of the star spikes
[/li]
[li]
Fix some color fringing on the stars by separating RGB channels and re-running star alignment with distortion correction enabled, and recombining with LRGB Combination.
[/li]
[li]
ICC Color Profile was added and the image saved as jpeg for viewing on the internet
[/li]
Comments