I think Kurt has the processing under control just fine from what I've seen, but here's my version regardless.
--- Auto Develop
Default values. To see what we got.
Focus is off a fair bit, a nasty yellow bias can be seen (totally correctable though!), some vignetting visible, severe oversampling (especially with the focus being out), a satellite trail visible, some stacking artifacts. There has been no dithering in between frames, which has resulted in 'streaking' pattern noise. The data also seems to be pre-color balanced (the light pollution appears to be 'correctly' colored) - this will add to noise.
--- Bin
To convert oversampling into noise reduction and undo debayering artifacts/interpolation.
Parameter [Scale] set to [(scale/noise reduction 25.00%)/(1600.00%)/(+4.00 bits)]
--- Crop
To get rid of stacking artifacts and to better frame the glob.
Parameter [X1] set to [15 pixels]
Parameter [Y1] set to [73 pixels]
Parameter [X2] set to [1265 pixels (-40)]
Parameter [Y2] set to [876 pixels (-12)]
--- Wipe
Vignetting preset.
Parameter [Dark Anomaly Filter] set to [5 pixels] in order to make sure no small dark anomalies can interfere with trying to detect the real background.
--- Auto Develop
Region Of Interest over glob's core.
Parameter [Ignore Fine Detail <] set to [3.7 pixels] in order to avoid optimizing for the fine noise.
Parameter [Outside ROI Influence] set to [5 %] in order to refrain from optimizing to much for anything outside of the glob's core (e.g. the Region Of Interest).
--- Deconvolution
I don't expect too much, but it's worth a try
Auto mask with default settings.
Nothing much happened, but that's also the cool thing about StarTools' Tracking feature and the Decon module (with Auto Mask)- it 'knows' when things would only get worse by applying decon (of course you can override this 'wisdom' if you think you know better!).
--- Band
By not pushing the background too much (e.g. by using a good ROI for AutoDev), this becomes almost unnecessary!
Parameter [Orientation] set to [Horizontal]
--- Life
We further push back noise by using the Isolate preset (using a full mask! e.g. Mask, Clear, Invert, Keep) which nicely isolates the glob from the backround. Also works great when you have larger scale nebulosity obscured by a busy star field.
--- Color
Final color calibration. The Color module comes up with something that is too green to my liking (this happens when the data has already been white balanced before it was given to StarTools, which will have exacerbated the noise in the red channel, which the Color module the over-compensates for). Fortunately color calibration by eye isn't too hard if you have a wide field like this - just make sure all star color temperatures are accounted for evenly (e.g. just as many red, orange, yellow, white and blue stars should be visible on average - ignore color in the glob though).
Parameter [Cap Green] set to [To Yellow]
Parameter [Dark Saturation] set to [3.40]
Parameter [Saturation Amount] set to [247 %]
Parameter [Green Bias Reduce] set to [1.49]
Parameter [Red Bias Reduce] set to [2.17]
All in-all, I''m not too thrilled with what I could come up with in terms of color for this image (star colors are close to correct though). It could be the pre-color balanced data or maybe some sort of LP filter was used?
--- Wavelet De-Noise
Final noise reduction;
Parameter [Scale 5] set to [90 %]
Parameter [Brightness Detail Loss] set to [10 %]
Parameter [Grain Size] set to [23.7 pixels]
Parameter [Read Noise Compensation] set to [3.83 %]
--- Develop
Purely a taste thing; I'm bringing back some Skyglow.
Parameter [Skyglow] set to [2 %]
-
- Autosave-restack.jpg (69.41 KiB) Viewed 4903 times
Processing and results will be much simpler with better data - it's actually a terribly unfair situation for newbies (they have to learn more/better tricks for post-processing, while they start off with worse data
).