Colour Help!

Questions and answers about processing in StarTools and how to accomplish certain tasks.
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Colour Help!

Post by thewormofautumn »

Hi all

Struggling to get the right colour balance with my image.

Camera: ASI2600MC
Filter: L-Extreme
Bortle: 5
Lights: 43
Darks: 40
Flats: 40
Dark Flats: 40

When I took these in NINA and applied the auto-stretch, it gave me an image that looked as follows. Obviously a little green, but definitely something I was expecting. There's a good separation of the different gases on display
Image
When I load the stacked FITS into StarTools (1.7.456) and get to the colour module, I just can't get anything like the above. It's way too "salmon", or most of the nebula is a greyish blue. Any help with this? and the entire nebula just looks mono-colour. I'd love to be able to change the colour balance and the hue (like in Photoshop) but don't get how it works here.

All of my stars are also bi-colour. The stars I see in a stretch in Photoshop for instance, are all multi-coloured.
Image
I've never got my head around the colour module in StarTools.

Cheers
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Re: Colour Help!

Post by thewormofautumn »

This is a 5min edit in Photoshop. I don't want to use PS because then I lose access to StarTools' excellent modules, but this is the sort of colour I'm going for. Very easy to achieve in PS.
Image
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour Help!

Post by admin »

Hi,

Would you be able to share the dataset so we can see what might be going on?

The L-Extreme filter is a tight O-III + Ha narrowband filter. Are you following basic procedures for narrowband (e.g. use the Compose module, use the Bi-Color modes, etc.)?

The L-Extreme filter should provide good separation that should be easy to process into a distinct bi-color. Neither the PS processing, nor the Nina stretch appears to produce good separations here, as neither are using color to show areas of Ha or O-III dominance.

If your stars are multi-colored in some other application, something is really wrong with your filter, dataset or processing application, as your filter simply does not permit you to record such coloring;
Image
Is your workflow including Wipe (never optional!)?

A good bi-color workflow that shows good separated emission dominance, should yield something like this for the keyhole area for example;
NGC3372_HaOIII.jpg
NGC3372_HaOIII.jpg (205.06 KiB) Viewed 3204 times
Ivo Jager
StarTools creator and astronomy enthusiast
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Re: Colour Help!

Post by thewormofautumn »

Hi there Ivo

Thanks for offering to take a look. Here's a link to the file straight out of DSS, on OneDrive.
If you need anything else, let me know and I'll add them to that directory.

You're also absolutely right on the star colour. That's my fault for missing that. Indeed, the stars are bi-colour in PS and even APP as well.

Lastly, this is probably the issue as I didn't follow the procedures for narrowband. I had a look at the Documentation, but can't see anything. Is there a link to anything (hopefully a video.. I hate reading :D)

Thanks!

https://1drv.ms/u/s!Ak9Pmua24KTPk4wsrCb ... Q?e=UteLzL
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour Help!

Post by admin »

Hi,

I'm not sure what might be going wrong, as it is trivial to show a nice Ha/O-III bi-color separation, even without special compositing and with a super basic workflow;

Open dataset, indicate it is linear.
  • AutoDev to see any issues
  • Crop away stacking artifacts.
  • Wipe, defaults
  • AutoDev for final stretch
  • Color module, Legacy preset.
Result;
Carina FITS 32bit.jpg
Carina FITS 32bit.jpg (444.84 KiB) Viewed 3202 times
Given you don't seem to be able to achieve this, can you tell me where things are going wrong for you?
Ivo Jager
StarTools creator and astronomy enthusiast
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour Help!

Post by admin »

For proper narrowband processing, use the Compose module to load the dataset as red, green and blue.
Set "Luminance, Color" to "L + Synthetic L From R(2xG)B, R(GB)(GB) (Bi-Color from OSC/DSLR)". See Compose module documentation for what these different modes do.

Just process the resulting luminance dataset as normal, bringing out detail. Chrominance (color) will be processed in parallel at the same time.

Once you hit the Color module, use the Bi-Color preset to get you started (or stay in Constancy mode). You can now freely, linearly throttle Ha and O-III using the red and green (or blue) bias respectively. Even cooler, you can choose from a huge number of popular bi-color palettes with a single click. Crucially, detail does not change; color and detail will remain completely independent.

All the above perks are courtesy of StarTools advanced signal evolution Tracking engine, making narrowband processing like this incredibly easy and extremely powerful.

Example of some nice Hubble-like golden yellows and blues;
StarTools_514.jpg
StarTools_514.jpg (296.66 KiB) Viewed 3200 times
Ivo Jager
StarTools creator and astronomy enthusiast
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Re: Colour Help!

Post by thewormofautumn »

Wow. How come I can't get that! :D

This is what I've been doing all morning, trying to keep it all to defaults so I can learn what everything does.

1. Open dataset and choose Linear, from OSC/DSLR with Bayer matrix and not white balanced
2. Autodev > no changes > keep (can't see any stacking artifacts? so no crop)
3. Wipe > no changes > keep
4. AutoDev again, choosing Redo global stretch > Keep
5. Decon > Auto-generate mask > Primary Radius change to around 2.4 > All > Keep
6. Sharp > Next > Auto-generate mask > Keep
7. Color > Fill Mask, which then gives me this:
Image
I clicked Legacy Preset per your example, and got this:
Image
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Re: Colour Help!

Post by thewormofautumn »

thanks for your post on the bi-color editing tutorial as well. I'd like to just nail this basic one first before I move onto that!
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Colour Help!

Post by admin »

Things are likely going wrong when you skip the cropping away of the stacking artifacts. There are actually multiple! Inspect the borders to see them.
DSS always leaves at least a one pixel border around the image (a long standing bug), but there is also a column of aberrant pixels a few pixels in from the left border. As a result of the stacking artifacts, Wipe backs off (as to not clip your signal) and removes very little of the bias signal. Cropping away the stacking artefacts should let me know if that lets you overcome the coloring issue.

Though not having an impact on coloring, I would also definitely bin as your dataset is oversampled.
Ivo Jager
StarTools creator and astronomy enthusiast
thewormofautumn
Posts: 39
Joined: Sun Nov 15, 2020 9:56 am

Re: Colour Help!

Post by thewormofautumn »

holy moley, you're right... I cropped, and then got this as the default in legacy mode. Wow
Image
I honestly couldn't see any stacking artefacts when I stretched the first time. Besides the one pixel border, how did you see the aberrant pixel column? I must be blind.
Post Reply