Combining RGB with Duoband for natural star colour

[thewormofautumn]

Hi folks

I'm starting a project on M8 at the moment. I will be using my 2600MC and L-Extreme to focus on the nebula and going for a muted Hubble-esque palette. However, I really never like the star colour that comes out of duoband/narrowband images so I want to mix in RGB stars. To do this, I will spend a night shooting without a filter and capturing full spectrum.

Question... how do I combine these two stacks when done so that only the RGB stars are applied? Is this possible?

I think I can do it in Photoshop once I've fully edited both stacks in StarTools but it would be nice to keep it all in StarTools.

Cheers

[Startrek]

I image with a 2600MC and L Extreme filter in my 8” and 6” newts and previously struggled with narrowband and Star colours. I also struggled with using the Compose module as well to achieve colour separation
All I do now is load my DSS stacked data set ( eg M16 ) in Startools v1.7 ( centre tab OSC linear data set ) process as normal in synthetic luminance or mono and just before I hit the colour module , create a good star mask , invert so stars are not sampled , select colour module , select bi colour preset at top, reduce green bias to reduce the green background , adjust red a bit , adjust bright and dark saturation a bit and I’m done
I use Shrink as well to tighten my Star field and adjust the mask after that
I end up getting a reasonable OSC narrowband image with white large stars and a mixture of blue and red tiny stars
See attached images of my M16 and M8
Hope the above helps and what your looking for ??

[thewormofautumn]

Hey Martin (it’s AdamJL from IIS)

Thanks for that. Not quite what I’m after unfortunately. You’re the master of natural looking nebula photos no doubt. I’m trying to go for a bi-colour nebula (e.g blue and red or orange/yellow and blue) but throw in the full gamut of RGB stars. If I just shot this target as broadband, I don’t think I’d be able to remix the colours of the nebula and keep the RGB stars (could be wrong?) so I’m going to try another way… I just don’t know what that way is yet

[almcl]

You might find this thread of interest:

viewtopic.php?f=4&t=2003&p=8798#p8798

About half way down Ivo mentions how to combine an RGB starfield with bi-colour data.

Is that what you are looking for?

[Burly]

v1.8 should address this issue using nbaccent module .

[thewormofautumn]

You might find this thread of interest:

viewtopic.php?f=4&t=2003&p=8798#p8798

About half way down Ivo mentions how to combine an RGB starfield with bi-colour data.

Is that what you are looking for?

that's exactly what I'm after thanks almcl!
I'll hopefully get to broadband imaging time tonight so I'll give this technique a go in the next couple of days. Cheers

v1.8 should address this issue using nbaccent module .

Oooo very interesting! StarTools going from strength to strength.

[Mike in Rancho]

I've had a rough time trying to pull this off myself. Not my data (don't have any like this), but some CN uploaded data for M16 that was taken in both SHO and RGB. Both processed fine by themselves, though I couldn't quite get the SHO to look like Hubble SHO no matter what matrix and settings I tried. Still, for the RGB I was able to make a nice stars only tiff after heal and layer subtraction.

But combining the RGB stars into the SHO image just hasn't worked. The color of fg method noted above turned out to be the closest, but still not a smooth enough transition to make it a satisfying result. Granted M16 is pretty hard for this, with many stars against a dark black of outer space, and many others in the midst of very bright nebulosity.

[thewormofautumn]

thanks for your insight, Mike. I'll give it a go and see what happens. Like you, I've had some issues with getting colours right for duoband, but this has helped me, hopefully it can help you too:

viewtopic.php?p=10263#p10263

I still don't entirely understand it, but I'll keep at it.

As a last resort, I know how to layer in Photoshop, so there's that route.

[Mike in Rancho]

Thanks I'll check that out more. For duoband "right" colors is sort of up for grabs. You can keep it red and teal, or just about anything else. Lots of options in the color matrix settings if you load it as OSC bicolor. Including some that sort of start looking SHO-Hubble-ish. Which is good for me. I'm okay with the various blends available, but not so much a fan of the faux SHO that strikes me as selecting and "photoshopping" a third color into existence.

That said, I just full spectrum modded a camera this week, and am getting first light right now through an LeNhance, so I'll have some data of my own to play with - though I have no RGB on this target.

The one I was mentioning earlier for blending RGB stars into SHO was a true S-H-O filters data, not duoband. Though I imagine the same tricks apply if wanting to add RGB stars. I just haven't found a great way to do it, even with all the features in ST layer module.

[Stefan B]

Hi Burly,

v1.8 should address this issue using nbaccent module .

How would the NB accent module do this? If understand the topic correctly @thewormofautumn wants to do a bicolor image, so he needs his NB data in R, G and B in the compose module and use the bicolor option. So only the NBaccent slot would be free for the RGB stars data. Is that working? Or would you load your broadband data in R, G and B and your NB data in NBaccent? But thus, you would not be able to do an HOO version (or any other bicolor matrix).

I would be really interested to know

@thewormofautumn: Together with Carles I recently discussed a bit of RGB/NB combination here, if you're interested:

viewtopic.php?f=11&t=2242

Best regards
Stefan