SharpCap Forums

Captured Sept - Oct 2025.

Data: SII 8x450s, Ha 6x600s, OIII 8x450s. 100 bias, 50 each darks, 50 each flats. 3h integration.


Equipment: Vixen 81s, Vixen 0.67 reducer/flattener, Celestron AVX mount, Atik 9-slot filter wheel, Altair filters (SII=6.5nm, Ha=7nm, OIII=6.5nm), Atik 314L+ CCD camera, QHY miniguider & QHY5lII-M guidecam.
Software Sharpcap, Pixinsight.

Processing (PixInsight):

• WBPP for calibration and integration

• PixelMath to combine the channels, red = (0.9*SII+0.1*Ha), green = (0.8*Ha+0.2*OIII), blue = (0.9*OIII+0.1*Ha) from Charles Bracken's book The Deep Sky Imaging Primer

• GradientCorrection

• ScreenTransferFunction + HistogramTransformation to delinearise

• CorrectMagentaStars script

• TGVdenoise

• RangeMask

• LocalHistogramEqualisation with inverted mask

• CurversTransformation, slight 'S' curve and saturation

I did try processing with Starnet++ to create 'starless' and 'starsonly' but didn't really make any difference.

The data stacks:


This is my first SII data - lots required.

For channel combination I also tried this formula I found in the PixInsight Forum:
R = (Oiii^~Oiii)*Sii + ~(Oiii^~Oiii)*Ha
G = ((Oiii*Ha)^~(Oiii*Ha))*Ha + ~((Oiii*Ha)^~(Oiii*Ha))*Oiii
B = Oiii

The Official PixInsight videos on Narrowband at https://www.youtube.com/watch?v=WYPVtmhXJYk were helpful.

My original plan was to exceed 10h of data but the weather .....

Dave

I like that colour scheme Dave with the SII combining into the red along with HA. It is a more naturalistic way than Hubble of combining the colours while using all of the data and it looks good. I might go back and explore redoing some of my stuff in that way.

But I don't understand how the PixMath equation you found is normalising the data so will have to read up

Tim

Tim

The formula was provided in the book but without any explanation. I did try a conventional ChannelCombination but the result was not as good (whatever good means with false colour?).

This is the HOO variant: Dave

Had another go at processing the HOO data.


These are the processing steps.

Dave

A slightly different processing route for SHO. SII and OIII had many stars, Ha not so. Combine SII-starless, Ha and OIII-starless instead of performing star reduction later.


Evaluation: no amount of processing wriggling can substitute for lack of data - 10h planned but only 3h captured. The amount of stars displayed felt 'about right'.

Dave

Hi Dave,
Have you tried the narrowband normalization process in PixInsight for mixing the channels I have found it works rather well it is what I used on the Soul image. Video of it below for info.

https://www.youtube.com/watch?v=n2h8jaAhsr4

Cheers
Nick

Thanks for that Nick, I will look into it.

Dave

Very useful Nick, thanks again. I got these with much less time spent on 'colours'. I went via the starless route with Starnet++, using Mode2 for the HOO variant.


Dave

Glad it was of use; I have been trying Starnet++ and a 30 day free trial of StarXterminator and think StarXterminator is just edging it at the moment but need to try it on a few more image. I will however definitely be buying PixInsight and BlurXterminator as a minimum.

Now if only there was something I could by to guarantee clear nights other than moving to a different country or continent.

Cheers
Nick