Two colour variations of NGC6888

[Topographic]

Attached are two different colour versions of the same image. The brown colour is my original version using the H(H+O)O preset. The other, redder version has been adjusted using gimp. I am interested to hear what people think. particularly from Ivo and the way colour is handled by other programs compared to StarTools. I see a lot of very red clouds on this target, i tried to emulate those in StarTools (experimental reasons only) and could not unless I used GIMP, which I rarely do.

[Mike in Rancho]

Neat! Make 'em bigger, though! Should be able to retain modest resolution and export to about 500kb size or so.

What were the alterations in Gimp? Selective mask? I see the surrounding Ha has been changed to deep red, but the Crescent itself seems to be retaining the H(H+O) for that emission.

[Topographic]

Neat! Make 'em bigger, though! Should be able to retain modest resolution and export to about 500kb size or so.

What were the alterations in Gimp? Selective mask? I see the surrounding Ha has been changed to deep red, but the Crescent itself seems to be retaining the H(H+O) for that emission.

To be honest my knowledge of using GIMP is pretty basic and my memory worse. I would think I just did a simple colour adjustment. For full resolution see https://www.astrobin.com/ofnh00/ Personally I like both images but I do wonder how extreme colour is on many images. This is taken with RASA 8, ZWO 183MM Pro and Astronomik MaxFR HA and OIII 12nm filters, 60 second subs.

[admin]

It's a little hard to comment on the coloring, without knowing exactly what happened in the GIMP.
If red H-alpha emissions are what you're after, then you'd want to map Ha to red and not any other channels. By currently mapping Ha to green as well, you're creating a yellow/brown where Ha is prevalent rather than red...

[fmeireso]

I don't want to hijack this thread but i am struggling with color in Startools too....

I always or mostly get pinkish, magenta like colors on emission nebulae.

I use Ha as Luminance from a mono DSLR (debayered sensor). I use a OSC color stack for color from an unmodded DSLR. I use L,RGB in compose.

So to get a reddish like emission nebulae i have to use GIMP, mostly but not always with some success.
I just guess my approach is wrong...

[Mike in Rancho]

Freddy I think pink from HaRGB is very common and also the expected result from such a combination.

NB Accent is probably the best way to go. Combining Ha into R may be a possibility, though just a try it and see sort of thing.

[fmeireso]

Yes Mike, that is probably true , but nonetheless very inaccurate coloring...

If you take an OSC , DSLR no filters no mods, and shoot the Heart eg for a couple of hours, and process it. It is faint , very faint but definately red.

Imho an unmodded cam give a good estimation what the color should be, allthough off couse the object is very faint. Maybe we should all forget about filters and tricks and shoot just plain OSC. 1 object a year...say 100 hours of integration time like ngatel did on CN with the squid. Result was just astounding...

I guess am kidding...

Actually i am experimenting with the layer module to get some more color in. But i find it hard to do the right thing and take the right steps...
I can't really blend the HA with the OSC color stack, because the OSC color seems to weak to get a red picture, i was however able to layer in more color from stars...i think there are possiblities but i need time to experiment....

[Mike in Rancho]

Yeah Freddy, perhaps. Although I guess the "accurates" and "should be's" are all dependent on what you are doing and what you are going for.

I tend to be skeptical when I see claims of "natural" color or appearance - which often seems to mean barely saturated into mild pastels. And any sort of wideband-narrowband combining can be riddled with assumptions and compromises, so you just have to make sure what you are doing has a good logic behind it.

A stock DSLR is sort of designed to mimic human eye response, though I think it could be too weak in deep reds like Ha, and maybe too strong in mid greens.

NB Accent may be the best way to combine things and force a color response that you find appropriate.

As to squids -- unless you can barely see mere parts of the blue outline, or the Ha bat is so throttled back that it is nearly invisible itself, I don't trust a single one, no matter how "amazing." PI mask tricks.

[admin]

I always or mostly get pinkish, magenta like colors on emission nebulae.

In the visual spectrum, that would be correct for most hydrogen-dominant emission nebulae.
The reason why this is correct, is that hydrogen ionization emissions are - by far - the most common (and strong).
Some people assume that, because H-alpha is the strongest, that therefore these emissions must be red. That would be an incorrect assumption.
The reason why this is incorrect, is because, there are other energy state transitions in the visual spectrum (aka the "Balmer series" of emission lines);

(from the linked page)

You are familiar with H-alpha (the red line), but the other lines (up to H-delta) are also present. These are all blue.
If you would take the colors you see above and blend them (as happens in real photos), you would end up with the pink/purple/magenta that you see in StarTools.

In other words, if you image in the visual spctum, and you end up with pinkish, magenta-like colors, then your image is scientifically correct and accurate.
Modifying color in some other application will only compromise this accuracy.

Hope this helps,

[fmeireso]

Ok Ivo, thanks for your explanation.

However this means then that actually most pictures we see of the eg Heart nebulae on the internet are just inacurate. Many of them are just red not pink.