NGC 2359 - Thor's Helmet
by Alejandro Tombolini
Introduction
Image of Jules publicated in Image Processing Challenges - Complete description of a narrowband image - Main notes: Carefull DynamicBackgroundExtraxtion and CanonBandingReduction to rescue fines details in SII wich was quite weak. Date: Feb 2013.
Processing
First a Crop of the three channels:
As I am using PixInsight 1.8 RC3 and do not have HistogramEqualization yet, I have used AdaptativeStretch applied to a clone of each channel to see and define background samples for DynamicBackgroundExtraction, generate the instance of DynamicBackgroundExtraction and then apply to the original channel. Same operation for SII and Hα.
It was not necesary DBE for OIII, but CanonBandingReduction to eliminate the horizontal line that can be seen in the base of the image. The values of the script as follow.
LRGBCombination R:SII, G:Ha, B:OIII.
Select a small preview and BackgroundNeutralization.
ColorCalibration. Same preview to be used as Reference image for Background Reference and the full image as Reference image for White reference. Note that I have unchecked the Structure Detection option.
DynamicPSF to generate a PSF to be used in Deconvolution and Starmask to protect the core of the bright stars. Then apply Deconvolution with the following parameters:
Consecutively twice MaskedStretchScript.
First 200 iterations, 0.1 Target median and Shadows clipping in the mask 0,01.
Second 200 iterations, 0.2 Target median and Shadows clipping in the mask 0,02.
Autozero Shadows in HistogramTransformation.
ATrousWaveletTransform for noise reduction using k-Sigma. (Juan, big, big improvement to have RTP for k-sigma noise thresholding, now this tool has no limits)
Clone SII, and apply HistogramTransformation and ATrousWaveletTransform to smooth and generate a mask to be used with CurvesTransformation.
Increase color and Saturation protecting bright areas.
Chrominance noise reduction with ACDNR without mask.
HDRMultiscaleTransform to increase details in bright areas, protecting stars with starmask.
Using the same Starmask to protect stars, apply LocalHistogramEqualization. Note that STF is active to visualize the process applied. The image in the left shows the actual stretch.
ATrousWaveletTransform to reduce noise and high light areas protecting bright parts of the nebulae with RangeMask.
Generate a new RangeMask over a clone of SII previously stretched with AdaptativeStretch and apply CurvesTransformation protecting bright areas.
Noise reduction to Chominance and Luminance protecting the nebulae with RangeMask.
HDRMultiscaleTransform twice, first 4 layers and finally 2 layers.
Transfer the STF to HT and apply.
To reduce stars, first generate a StarMask and apply MorphologicalSelection protecting background.
Second, with the same starmask, adjust curves and green on stars.
New starmask to protect stars and increase details in the nebulae with UnsharpMask.
Last noise reduction using MultiscaleMedianTransform tool in layers 1 and 2.
Final Image:
