IKO NGC6888 - Processing Competition

[matt_baker]

Here's my modified HOO attempt. Good luck to everyone, there's already so many stunning renditions!

Processing done in Pixinsight.

Here are my brief steps that I took for Ha and OIII:

• Deconvolution StarMask
• Extract Lum, stretch and create range mask for highest areas of signal
• Get PSF using PSFImage script
• Applied 20 iterations of deconvolution
• Stretched Lum out to 75% along the histogram for an MLT mask
• Applied a subtle amount of MLT to both filters

Non Linear Stage:

• Stretched using HistogramTransformation
• Used Jon Rista's masks method for TGV for both filters
• Applied small amount of TGV to Ha and a little more to OIII
• PixelMath combination of R = Ha, G = OIII*0.5+Ha*0.5, B = OIII
• Starnet++ to get rid of stars
• Curves transformation for saturation and overall brightness
• ColourMask for the blues
• RangeMask and CloneStamp to get rid of residual background and only exposing the crescent itself
• Curves just for the blue outer shell
• ExponentialTransform to brighten nebula
• LocalHistogramEqualization for constrast
• UnsharpMask for more crispiness
• DarkStructureEnchance
• StarMask and MorphologicalTransformation for smaller stars

 

 

Edited September 29, 2020 by matt_baker

[Adam1234]

Here's my attempt, based on HOO, R = 75%Ha, 25%Sii; G = 85%Oiii, 15%Ha; B = 100%Oiii

Processed in photoshop. Curves, levels, vibrance, astronomy tools action set etc, stretched Ha separately and used as luminance.

 

 

Adam

Edited September 29, 2020 by Adam1234

[Seelive]

My first real attempt at narrow band processing since I only use a DSLR for my own astro-photographs.  Processed using IRIS and Starnet++.

1.       Combine data as SHO

2.       Equalise and offset sky background histograms peaks

3.       Initial small stretch and remove stars using Starnet++

4.       Create stars only image

5.       Apply curves to nebula image to equalise and offset histograms

6.       Blend in stars only image

7.       Apply minor RGB colour correction

8.       Small ASINH stretch

9.       Light wavelet filtering

10.     Small application of CRISP filter

The Ha still seems to be a bit too strong for my liking so I may try an alternative colour pallet.

 

[TerryMcK]

Here is my effort.

Processed in PixInsight first of all

• Channel Combination RGB-HOO
• Then NBRGB Combination adding SII into B Channel
• ABE
• ColorCalibration
• Multiscale Linear Transform
• Histogram Transform
• Curves Transform
• Local HistogramEQ
• Invert
• SCNR red .36
• Invert
• StarMask
• SCNR to remove green on star mask
• Invert StarMask
• SCNR to remove more green on star mask
• Invert StarMask
• Pixelmath to add the two images together

Finished off with Photoshop

• Minor level balance
• Reduce stars 3 times
• Invert image to colour replace/reduce some of the blue rings around stars
• Invert image right way around again
• Sharpen

 

Edited October 1, 2020 by TerryMcK

[A320Flyer]

I just had to have another go at pulling out the Oiii

Bill

[almcl]

Wonderful data again!  But found this really hard to process, particularly without losing some of the Oiii data. 

 

Processed in StarTools 1.7.421 alpha standard work flow:

--- Compose   >  Auto Develop >  Crop >  Wipe > Auto Develop >  Contrast > HDR >  Wavelet Sharpen > SNR-aware Wavelet Sharpening > Deconvolution > Shrink >  Life >  
Color
----Parameter [Matrix] set to [SHO 40SII+60Ha,70Ha+30OIII,100OIII]

> Wavelet De-Noise
 

Edited October 3, 2020 by almcl
new image version

[MartinB]

Been away out of internet reach for the past few days so a bit of a last minute entry.  Great to see the range of different processes and like many I have found this one tricky to say the least.  I've gone for a simple bicolour Ha and OIII process in the end mapping the OIII to blue and green with more weight on the blue.  This was all processed in PS other than use of Starnet via PI.  I did try deconvolution in PI but wasn't happy with the appearance of the nebula afterwards, especially those pesky bright Ha regions.

 

[skyrunner]

Lovely data, great to have such low noise O3 & S2 as starting points. Processed in Pixinsight throughout.

Luminance processing:

- Max(Ha, S2, O3) -> Lum

- Deconvolution to sharpen stars

- TGV denoise with Luminance mask to reduce background noise

- Histogram stretch (multiple)

- MMT to shrink stars

- Starnet to create starless_lum image

- Blend 70% lum with 30% starless_lum

- Sharpen with Local Histogram Equalisation & MMT using starless_lum as nebula mask

Colour processing:

- LinearFit O3, S2 to Ha

- RGB Combine (Ha, 5%Ha+25%S2+70%O3,O3)  [Combinations evaluated using SHO-AIP script]

- BackgroundNeutralisation

- Histogram stretch (multiple)

- Curves transform

- LRGB Combination of lum and colour 

- Curves transformation

- Remove yellow halos on big stars with colour mask

I am also posting the starless monochrome image - I was blown away when I saw it and think it makes a great image on its own.

 

 

 

 

Gordon

Televue-85, WO RedCat 51, LX200 10"

HEQ5-Pro, Star Adventurer Pro

ZWO ASI183MM Pro, Canon Rebel T3i Astro modded

Pixinsight, SG Pro, Affinity Photo, PHD2

 

Edited October 3, 2020 by skyrunner

[newbie alert]

Thanks to FLO for the data, fantastic to work with..heres my rendition, i had some issues last week as i processed it last week, only to find my photoshop wasnt saving the file, and my new also wasnt working.. adobe said they could fix it but possibly lose the work i wanted to save..I did lose the file, and no matter how many times i tried to redo it i never came back with the same result..so its somewhere where i wanted it but not exactly where it was...Again its fun to play with, learning as you go.. processed in PS ,red ha, with 20% S11, green O111 WITH 50% HA, BLUE 100%  O111, Levels curves and Carbonie tools 

 

 

[Rob63]

A huge thank you to to FLO & IKO for this, it's a rare treat to work with such fantastic data especially in OIII & Sii.

I wanted this image to show the magnificent background Ha nebulosity with an RGB look while also using the SHO data to highlight the crescent 'bubble' in a complimentary colour so that it stands out while still looking natural.
To achieve this I created two versions (plus a third for the stars) and blended them together.
Full processing details are listed below the image.

Pixinsight

load Ha, OIII, SII
Crop
DBE
slight Noise reduction - MultiscaleMedianTransform  (MMT)

Ha, OIII, SII => stacked as Master Lum
Crop
DBE
slight Noise reduction - MMT
Deconvolution
Histogram Transformation
Masked stretch

Channels combined with PixelMath, I created three versions, detailed below. 

1st image - Background Nebulosity
For the background: The goal was to keep this as an RGB style rich red to show off the Ha nebulosity.
Histogram Transformation
Masked stretch
SCNR
LRGB combined with master lum
Starnet to remove stars and process separately (N.B. lowered stride value to avoid artifacts)

2nd image for crescent 'bubble'. 
The goal here was to create an SHO version to provide a complimentary colour so that the crescent stands out (but doesnt offend the eye). Chose a "hubble style" blend to provide yellows/browns/golds and preserve the OIII shell as blue.
PixelMath
Red: Sii*0.35+Ha*0.65
Green: Ha*0.35+Oiii*0.65
Blue: OIII
Histogram Transformation
Masked stretch
SCNR
LRGB combined with master lum
HDR Multiscale Median Transform to improve detail
Starnet to remove stars and process separately (N.B. lowered stride value to avoid artifacts)

3rd image for stars (more RGB like)
RGB mapped as SII Ha OIII
Several gentle applications of Histogram Transformation
Starnet to extract stars and process separately
Photometric color calibration for background neutralisation (narrowband mode)
SCNR
Curves (saturation)
Star reduction - Morphological Transform

Export nebula and crescent to photoshop
Blend images meet goals outlined above
Adjust colour, saturation & contrast
Check for starnet artifacts around brighter stars - fixed a couple of bright star halo artifacts with clone stamp 

Back to Pi
Local Histogram Equalisation
SCNR
Exponential Transformation (PiP)

Export back to photoshop
Add stars layer (linear dodge(add) ) over the "background and crescent" layer
Any final tweaks

Merge and save
 

Rob.

[Neko]

Ok...so this is my  second/last try I guess.. It is quite different than the previous one

I wanted to get more of the Oiii outer "shell " of the crescent, and also the dark nebulas on the upper right corner.

is in SHO and processed in StarTools 1.7.422 GPU version. Then slightly touched in windows "photo" editor...

Loaded the three files first and saved it as a single Luminance file in Mono on compose module. 

Also loaded H-alpha and Oiii individually and stretched a bit. 

Then loaded that luminance in the Red channel, H-alpha in Green, and Oiii in Blue. 

then as follows...

--- Compose
Parameter [Luminance, Color] set to [L + Synthetic L From RGB, RGB]
Parameter set to [On]
Parameter [Lum Total Exposure] set to [Not set]
Parameter [Blue Total Exposure] set to [2h52m (172m) (10320s)]
Parameter [Green Total Exposure] set to [1h46m (106m) (6360s)]
Parameter [Red Total Exposure] set to [2h47m (167m) (10020s)]
Image size is 2750 x 2200
Type of Data: Linear, was not Bayered, or was Bayered + white balanced
--- HDR
Parameter [Small Detail Precision] set to [Max]
Parameter [Channels] set to [Brightness Only]
Parameter [Algorithm] set to [Reveal All]
Parameter [Dark/Bright Response] set to [1.00]
Parameter [Detail Size Range] set to [1000 pixels]
Parameter [Strength] set to [1.3]
--- Wavelet Sharpen
Parameter [Structure Size] set to [Large]
Mask used (BASE64 PNG encoded)
--- SNR-aware Wavelet Sharpening
Parameter [Scale 1] set to [100 %]
Parameter [Scale 2] set to [100 %]
Parameter [Scale 3] set to [100 %]
Parameter [Scale 4] set to [100 %]
Parameter [Scale 5] set to [100 %]
Parameter [Mask Fuzz] set to [4 pixels]
Parameter [Amount] set to [300 %]
Parameter [High SNR Size Bias] set to [85 %]
Parameter [Low SNR Size Bias] set to [99 %]
Parameter [Dark/Light Enhance] set to [50% / 50%]
--- Deconvolution
Parameter [Image Type] set to [Deep Space]
Parameter [Secondary PSF] set to [Off (Primary Only)]
Parameter [Primary PSF] set to [Moffat Beta=4.765 (Trujillo)]
Parameter [Primary Radius] set to [1.5 pixels]
Parameter [Iterations] set to [11]
Parameter [Error Diffusion] set to [Off]
Parameter [Mask Fuzz] set to [8.0 pixels]
Parameter [Deringing] set to [50 %]
Mask used (BASE64 PNG encoded)

--- Shrink
Parameter [Mode] set to [Tighten]
Parameter [Halo Extend] set to [1 pixels]
Parameter [Iterations] set to [15]
Parameter [Regularization] set to [0.80]
Parameter [De-ringing] set to [5 pixels]
Parameter [UNKNOWN] set to [1.5 pixels]
--- Crop
Parameter [X1] set to [18 pixels]
Parameter [Y1] set to [0 pixels]
Parameter [X2] set to [2750 pixels (-0)]
Parameter [Y2] set to [2200 pixels (-0)]
Image size is 2732 x 2200
--- Color
Parameter [Bias Slider Mode] set to [Sliders Reduce Color Bias]
Parameter [Style] set to [Scientific (Color Constancy)]
Parameter [LRGB Method Emulation] set to [Straight CIELab Luminance Retention]
Parameter [Matrix] set to [SHO 50SII+50Ha,50Ha+50OIII,100OIII]
Parameter [Dark Saturation] set to [5.4]
Parameter [Bright Saturation] set to [5.3]
Parameter [Saturation Amount] set to [200 %]
Parameter [Blue Bias Reduce] set to [2.29]
Parameter [Green Bias Reduce] set to [1.16]
Parameter [Red Bias Reduce] set to [5.68]
Parameter [Mask Fuzz] set to [1.0 pixels]
Parameter [Cap Green] set to [9 %]
Parameter [Highlight Repair] set to [10 pixels]
--- Life
Parameter [Detail Preservation] set to [Linear Brightness Mask]
Parameter [Compositing Algorithm] set to [Multiply, 2x Gamma Correct]
Parameter [Inherit Brightness, Color] set to [Off]
Parameter [Output Glow Only] set to [No]
Parameter [Airy Disk Sampling] set to [128 x 128 pixels]
Parameter [Airy Disk Radius] set to [8 pixels]
Parameter [Glow Threshold] set to [3 %]
Parameter [Detail Preservation Radius] set to [20.0 pixels]
Parameter [Saturation] set to [75 %]
Parameter [Strength] set to [27 %]
Parameter [Mask Fuzz] set to [1.0 pixels]
--- Entropy-driven Detail Enhancement
Parameter [Resolution] set to [Medium]
Parameter [Channel Selection] set to [Red (SHO SII,Vis.Ha)]
Parameter [Strength] set to [65 %]
Parameter [Midtone Pull Filter] set to [20.0 pixels]
Parameter [Midtone Pull Strength] set to [50 %]
Parameter [Dark/Light Enhance] set to [0% / 100%]
File saved 
--- HDR
Parameter [Small Detail Precision] set to [Max]
Parameter [Channels] set to [Brightness Only]
Parameter [Algorithm] set to [Reveal All]
Parameter [Dark/Bright Response] set to [1.00]
Parameter [Detail Size Range] set to [1000 pixels]
Parameter [Strength] set to [1.3]
Undo
Image size is 2732 x 2200
Redo
Image size is 2732 x 2200
Undo
Image size is 2732 x 2200
Redo
Image size is 2732 x 2200
--- Wavelet De-Noise
Parameter [Grain Size] set to [4.1 pixels]
Parameter [Walking Noise Size] set to [1.0 pixels]
Parameter [Walking Noise Angle] set to [0]
--- Wavelet De-Noise
Parameter [Scale 1] set to [95 %]
Parameter [Scale 2] set to [95 %]
Parameter [Scale 3] set to [95 %]
Parameter [Scale 4] set to [95 %]
Parameter [Scale 5] set to [50 %]
Parameter [UNKNOWN] set to [4 %]
Parameter [Scale Correlation] set to [50 %]
Parameter set to [50 %]
Parameter [Uniform Grain Retention] set to [50 %]
Parameter [Grain Dispersion] set to [1.9 pixels]
File saved 
--- Develop
Parameter [White Calibration] set to [Use Stars]
Parameter [Gamma] set to [0.87]
Parameter [Skyglow] set to [4 %]
Parameter [Digital Development] set to [Off]
Parameter [Blue Luminance Contrib.] set to [136 %]
Parameter [Green Luminance Contrib.] set to [100 %]
Parameter [Red Luminance Contrib.] set to [100 %]
Parameter [Dark Anomaly Headroom] set to [5 %]
Parameter [Dark Anomaly Filter] set to [2.0 pixels]
U
--- Wavelet De-Noise
Parameter [Grain Size] set to [2.0 pixels]
Parameter [Walking Noise Size] set to [1.0 pixels]
Parameter [Walking Noise Angle] set to [0]
 

 

[x6gas]

Well once again thanks to the Ian King Observatory and to FLO for supplying really wonderful data - this data is incredible, in fact.  It's great to have have such wonderful data as you can remind yourself of all sorts of techniques and tricks instead of spending the time wrestling the data to submission.

I offer The Crescent Nebula three ways.  For each version, I deconvolved the data in PI and applied a gentle dose of Multiscale Linear Transform Noise Reduction before moving to PhotoShop.  I also used a starless version of the OIII (generated using Starnet ++) to give that channel a bit of a boost.  Once in PI, there was a bit of sharpening using High Pass filter at various strengths but most of the "processing", such that was needed, was playing around with the relative strengths of each channel and some playing with selective colour.

There are some absolutely brilliant renditions of this target on this thread.  I've never been completely happy with the versions that I've produced using my own data but I like what I've done with this data much more.

The first is a relatively straightforward bicolour version.  This has an additional OIII layer of the nebula added to bring out the fainter stuff: 

And this was supposed to be a gentle SHO version, but it actually looks a bit racey to me now that I see here...

and finally an SHO where I have pushed the Ha a bit harder:

 

Thanks again IKO and FLO!

Ian

 

Edited October 4, 2020 by x6gas

[alacant]

Hi everyone.

Didn't know there was another processing competition. I usually never get beyond the beginners' section of the forum. Maybe this is too late...

Not much monochrome so...

StarTools 1.7 using compose bi-colour with red = S+H. Stars tamed by the new tighten module; works wonders with the right mask. Colour and make-it-look-nice by DarkTable. Ubuntu 20.04.

Clear skies and stay safe.

 

 

Edited October 5, 2020 by alacant

[bottletopburly]

replaying with data using Startools1.8.506

 

[dannybgoode]

Whilst by no means a competent image processer I do love playing with this IKI data.  So clean and just so much in there.  Way better than anything I could capture, particularly under my Bortle 100000000 skies!

So here is my attempt.  Initial processing in PixInsight; ABE, colour integration (SHO), EZ_Denoise, EZ_starreduction, EZ_HDR, bit of a tweak here and there with saturation etc.  Nothing fancy, mainly because I do not fully understand PI yet however I am getting there.  

Then into DxO PhotoLabs 3 to have a play with the colour wheel and bring down the saturation as it was overly red and green and also into Affinity Photo to play around with the levels there and then merged the two files together to give the final result.  

 

Slight edit - prefer this version

 

 

Edited September 24, 2021 by dannybgoode

[dannybgoode]

Thought I would have a second stab at this and go for something completely different.  Yes, AP can be all about the colour but what if we removed that distraction to focus on the form and structure and process the image in a more art-like fashion and go black and white, chuck in some false vignette, perhaps add a border.

You would end up with something like this.  This target struck me as an ideal candidate for such processing due to the clouds of nebulosity surrounding a very interesting and intricate core object

[AstroAndy]

Here is my work. This looks like the cleanest data I worked with so far. It's also my first HOO biclolor composition. First, the different channels, then 2 color renderings.

All in photoshop, triplicated layers to work separately on stars and DSO, and that was pretty much it.

[AstroAndy]

And here are the color ones.