Does anyone have any general tips for getting good dual band (HOO) color?

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
xonefs
Posts: 55
Joined: Sun Jul 18, 2021 7:30 am

Does anyone have any general tips for getting good dual band (HOO) color?

Post by xonefs »

This is not a specific troubleshooting since I haven't got any fresh on my mind, but every time I try to process HOO data I can't get good color separation (with the exception of veil nebula).

What usually happens is the data will initially be very pink dominant, then if I try to adjust colors in color module to get some separation and blue to come out the rest of the image starts looking weird and other parts of nebulosity start to lose color.

I tried to process carina in HOO the other week, and to try to get any color separation the edges of the nebulosity started to turn washed out and gray.

Then this was a while ago and I already asked about it, but trying to process soul nebula and bring out oiii results in the edges of the nebulosity getting washed out and it still not looking quite right. And that's not getting into the hues I would be looking for.

I really like a lot of HOO images I see with red and then shifting to purple->blue.
Image
Image

so has anyone processed good HOO data and managed to avoid that and get the kind of colors I'm talking about? (other than supernova remnants)

I would like to see some examples
Mike in Rancho
Posts: 295
Joined: Sun Jun 20, 2021 10:05 pm

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by Mike in Rancho »

Nice images. And yeah this does sound sort of familiar to a prior conversation around here?

I believe I know what you are talking about. Sometimes images are posted that have significantly strong OIII, as if they were deep and almost unbelievably smooth pools of blue. And of course the separation you are talking about.

I remain skeptical. Particularly if it is explained. Some of these things I think are done by "fitting" the channels to each other. My guess is that re-weights the contributions of the source elements. In ST, some weighting can be done with the Synth L in Compose, but color itself I don't think so?

There's a thread here where Ivo has a nice graphical continuum that shows just how a bicolor balance works, and saturates, and thus what it means when regions start becoming whiter as far as the bicolor balance goes.

There's also the matter of picking the right targets. Some just don't have strong enough OIII to warrant the images that seem to result. I find the Soul to be one of those, whereby you really have to balance back the Ha to start showing some blue -- and even then it's not the full blue central pools often seen on the internet. To me the Soul tends to be a rather broken and spotty target. But then again for one I've only tried it once, with limited integration, and an L-eNhance rather than real mono OIII. I also obviously don't do any range mask hypersmoothing or whatever seems to be the latest rage. :roll:

Here's an HOO I just did last night. Good data and a seemingly amenable target. This is Eyal's "tutorial" data of NGC2070 from the big Elf processing thread on CN (might still be linked from the main ST website). It's actually SHO data and I did it that way previously, but decided to see how it would look in HOO. The original file origins and weighting are a bit of a mystery, but nonetheless I actually had to pull back the OIII a bit in order to reveal some more of the red. Also not too many areas turned to a foggy white, though again that might say more about the actual regions of gas concentration than anything else?

Smaller compressed version, there's a bigger 1600 pixel one on CN but even that is scaled down.

Eyal NGC2070 ST8 HOO 1B 1280.jpg
Eyal NGC2070 ST8 HOO 1B 1280.jpg (471.55 KiB) Viewed 167 times

One thing you might be able to try in your datasets, based on that continuum of Ivo's, would be to alter the saturation (changing the span of the "white zone"), perhaps both overall and as to light and dark, and see if that changes the fading fog a bit?
xonefs
Posts: 55
Joined: Sun Jul 18, 2021 7:30 am

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by xonefs »

Which thread is that on the white in bicolor?

Yeah I think you may be right on the soul/heart being overprocessed for that oiii

Carina has a lot of oiii though and my data looked like I got a lot, though less than normal since I shot it at 5* elevation and there was some atmospheric extinction, but it still seemed bright.

But I can’t get any of that purple/blue to come through (I changed the color for the one I posted but I tried classic hoo too)
Mike in Rancho
Posts: 295
Joined: Sun Jun 20, 2021 10:05 pm

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by Mike in Rancho »

Hmm, let me see if it lets me do this....Ivo's post and graphic are here: viewtopic.php?p=11013#p11013

I refer back to it every now and then as it is a good explanation, and the example is tailored for HOO. But yes the meaning of white is described, as well as what you should see when changing saturation, and altering the H / O bias.

That Carina looks amazing, good job! And feel free to post the stack to see what the ST crowd here can yank out of it for the blue channel. ;)

And you did it at 5° to boot? I just looked it up, and unless the atmosphere can bend the light rays enough for me to see below the horizon, this one is not going to be available to me at 34N.

Last year I reached down and scraped Omega Centauri off the horizon, 8.5° alt at transit. Mostly just a test, and for the one part of my back yard that has a semi-low open window to the south, I can't see Polaris so have to drift align the PA. The big glob is also one of the brightest things in the sky. I'll try it again this year when it comes into position, but I have some other ideas in mind for more Southern Hemi low-riding too. :D
Rkonrad
Posts: 218
Joined: Sun Feb 07, 2016 2:55 am

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by Rkonrad »

Just to put in my 2 cents worth, personally picking the right target is important as there should be at least a certain portion of Oiii there. Two examples are the Rosette nebula https://flic.kr/p/2jMt7rs which has some but not a lot of Oiii and M42 https://flic.kr/p/2mE2yVT

Generally, I usually push the dark saturation fairly high and reduce the green (a lot). This tends to make the red from the Ha look more natural.
I hope that helps.
Richard
Mike in Rancho
Posts: 295
Joined: Sun Jun 20, 2021 10:05 pm

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by Mike in Rancho »

As a bit of an interesting follow-up, I just processed some SHO tutorial data on the Soul the other night, and apart from the tricolor NB it looks very much like the first image above.

I'm pretty sure I've seen this before on Soul, and so believe it is probably target dependent, as in, this is what the Soul nebula is. So, if one wants to show off the blue in the main target, and there is blue or whitish (more equal concentrations per that chart) in the region, it is going to be enhanced and revealed as well.

I think I've also seen something like this on the Crescent?

Of course there are more than a few versions of Soul floating out there which have a strongly-enhanced blue center, yet no blue or gray at the edges or scattered about the periphery. That's starting to make me skeptical of the processing. Unless there's something I'm not thinking of? :think:
xonefs
Posts: 55
Joined: Sun Jul 18, 2021 7:30 am

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by xonefs »

Mike in Rancho wrote: Sat Jan 22, 2022 4:12 am As a bit of an interesting follow-up, I just processed some SHO tutorial data on the Soul the other night, and apart from the tricolor NB it looks very much like the first image above.

I'm pretty sure I've seen this before on Soul, and so believe it is probably target dependent, as in, this is what the Soul nebula is. So, if one wants to show off the blue in the main target, and there is blue or whitish (more equal concentrations per that chart) in the region, it is going to be enhanced and revealed as well.

I think I've also seen something like this on the Crescent?

Of course there are more than a few versions of Soul floating out there which have a strongly-enhanced blue center, yet no blue or gray at the edges or scattered about the periphery. That's starting to make me skeptical of the processing. Unless there's something I'm not thinking of? :think:
yeah I think this comes down to a lot of heavyhanded processing with masks basically painting blue on skewing our perception of what's normal for certain targets.

though I'll admit there are plenty of examples where it looks tastefully done and you can't tell. I would like to try both and do a realistic and 'artistic' version sometime and present them as such, but I guess that would be outside of startools.
fmeireso
Posts: 77
Joined: Mon Sep 28, 2020 8:46 pm
Location: Belgium

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by fmeireso »

I try to process HOO too. I use mostly an L-enhance filter and use (when i can) a mono modded DSLR with Ha filter, thatone used as Luminance to boost the HA. Below imgage is rather strong colored. In most cases the OIII goes the cyan blue, wich I alter in GIMP.

Often in Startools is see pink coming on, especially when ones tries the entropy, i avoid it , but in then Superstructure i need the saturation part alot.

Dunno if you like it. Must say i have quite some pattern noise, probalby by not enough dithering in my mono camera. I am just starting this approach, i am not quite there yet....

This thread is about this too...
viewtopic.php?f=10&t=2370
Attachments
SoulNew10 (Large).jpg
SoulNew10 (Large).jpg (356.46 KiB) Viewed 35 times
Mike in Rancho
Posts: 295
Joined: Sun Jun 20, 2021 10:05 pm

Re: Does anyone have any general tips for getting good dual band (HOO) color?

Post by Mike in Rancho »

There may be some merit to that, Freddy.

For artistic purposes and without drawing masks, I think there are a number of things that can be done, probably in conjunction with each other, if one doesn't like the blue-gray edge fog that can result from throttling back Ha a lot in order to reveal OIII.

So after finishing and turning tracking off (some may work during tracking, would have to experiment), one can go back and forth between things like FilmDev, where you can lower the overall gamma, or even reduce the L contribution of say the blue channel, and SS where you can use the brighten preset to bring things back, but just what are deemed the super structures. Possibly a lot of it could be done in SS alone, as there are quite a few controls. Filter module may also help some, though it would probably take a reject rather than a fringe kill to make any real difference on the gray fog.

These things could increase contrast and darken the background quite a bit, though.
Post Reply