Soul Nebula-Hoo

[Burly]

First try at the soul Nebula ,zwo533mc L Extreme filter ,6.5 hrs approx bortle 5,stacking in siril extract o111,Ha script ,using Bi color from osc option not sure if that was the correct option loading as HOO ,Ha as L ,if anyone would like to have a go feel free
Ha/O3 https://drive.google.com/file/d/1jsm1Eb ... sp=sharing
HA/ https://drive.google.com/file/d/1fj9qn7 ... sp=sharing
O3/ https://drive.google.com/file/d/1i3KUfS ... sp=sharing

[admin]

You might want to not use Siril's separate O-III and Ha extraction, and let StarTools handle this instead.

Use the Compose module to load the same (color/3-channel) stack in red, in green and also in blue. Then set Luminance, Color to "L + Synthetic L From R(2xG)B, R(GB)(GB) (Bi-Color from OSC/DSLR)".

This should make the most of the signal you captured. Once you made it to the Color module, click the Bi-Color preset to get you started (use the Matrix option for alternative bi-color mappings).

The Soul nebula is a little bit of a tricky one, as there is not heaps of O-III signal to be had (it's just the nature of the object). The image you present here is indeed a good representation of what to expect in terms of color. You may be able to eke out some patches of blue in the central nebula by throttling the Ha (red bias) and boosting O-III (blue or green bias, when bi-color matrix is active).

If the blue stars take over too much for your taste (unfortunately a fact of life when doing Ha/O-III imaging), you could use the highlight repair slider to reduce their coloring if you wish. Using the Shrink module's Color Taming can also be used for this purpose.

Hope this helps!

[Burly]

Thank you ivo for clarification on that ivo much appreciated I do have the integrated file from Siril too which is what I usually use I will link that file too Do i leave the L channel blank and not load the L extreme data into lum channel?.

You might want to not use Siril's separate O-III and Ha extraction, and let StarTools handle this instead.

Use the Compose module to load the same (color/3-channel) stack in red, in green and also in blue. Then set Luminance, Color to "L + Synthetic L From R(2xG)B, R(GB)(GB) (Bi-Color from OSC/DSLR)".

This should make the most of the signal you captured. Once you made it to the Color module, click the Bi-Color preset to get you started (use the Matrix option for alternative bi-color mappings).

The Soul nebula is a little bit of a tricky one, as there is not heaps of O-III signal to be had (it's just the nature of the object). The image you present here is indeed a good representation of what to expect in terms of color. You may be able to eke out some patches of blue in the central nebula by throttling the Ha (red bias) and boosting O-III (blue or green bias, when bi-color matrix is active).

If the blue stars take over too much for your taste (unfortunately a fact of life when doing Ha/O-III imaging), you could use the highlight repair slider to reduce their coloring if you wish. Using the Shrink module's Color Taming can also be used for this purpose.

Hope this helps!

[Burly]

Would this be a more representative process ivo loaded as above not loaded as lum , reduced red bias and upped R,B to align channels roughly the same , i left the hint of green stars as is though i did try using star colour from a different colour matrix using the layer module and that works quite well,cheating i know

[Mike in Rancho]

Hi Burly,

I tried the download last night but it didn't work, I think maybe you were in the midst of changing up the links? They seem to be all working now so I might try again.

For the combined L-eXtreme file, I would not use it in L. That would just turn that OSC file into an overall luminance, but compose is already creating the same thing (the Synthetic L) from the R, G, and B - with Gx2 if you select that - as extracted from the OSC file. Really it would just be a matter of weighting I guess, but you can do that anyway with the channel exposure settings. So perhaps superfluous or a more complicated and confusing way of getting to the same place. But if you needed to calculate it out, I believe the baseline is that the L channel is weighted 50% against 50% for the total of the R, G, and B channels.

My minimal Soul experience, mostly with others' data as I only have a small amount of my own, matches Ivo's conclusion of course. The OIII is somewhat weak, so bringing it out takes a lot of Ha throttling. It also tends to be noisier, which can be disheartening. Mostly though, the throttle back of the relative concentrations can leave the periphery with a gray, ghosted appearance that is usually not appreciated.

Playing with saturation may help, but I get the feeling a lot of the heavy red+blue Soul images out there may be somewhat manipulated as to that balancing. Particularly from duobands which would acquire the same integration time on both bandpasses.

Your files might be an interesting comparison though on Siril's extract functions. I recall discussions about it on CN perhaps a year ago. The alleged "selling point" was that Siril was superpixeling the separate Ha (R) and OIII (unsure but assumed B+G) pixels, and then drizzling the resultant files back to size, avoiding debayering. While I think debayering is getting too much of a bad rap, it was recently posted that the so-called drizzling isn't really drizzling. Siril is just straight resampling it 2x to re-scale back to the original size.

So side-by-side you might be able to see whether one method gives finer or cleaner detail?

[Burly]

Hi Mike yes I had file access set wrong now sorted , look forward to your take , also being osc you need more integration time compared to mono, if I can get another 6hrs then I’d be happier, though uk weather not very good at the moment.

[admin]

Would this be a more representative process ivo loaded as above not loaded as lum , reduced red bias and upped R,B to align channels roughly the same , i left the hint of green stars as is though i did try using star colour from a different colour matrix using the layer module and that works quite well,cheating i know

Textbook! A rendering that is precisely as expected with the O-III starting to color the central region that ghostly teal/blue.
(indeed, as Mike already expertly commented, you don't load anything for luminance, as StarTools already creates a properly weighted synthetic luminance dataset for you)

[Burly]

Thanks Ivo

[dx_ron]

Nice image, Burly! I, too, am trying to improve my duoband processing (because that's most of what I am collecting right now).

I don't follow what you meant by "upped R,B to align channels roughly the same" - could you expand on that?

[Burly]

Nice image, Burly! I, too, am trying to improve my duoband processing (because that's most of what I am collecting right now).

I don't follow what you meant by "upped R,B to align channels roughly the same" - could you expand on that?

In colour module use the sliders to for RGB to increase or decrease bias to suit ron , then used the left hand graph showing channels to roughly be the same , hope that helps .

Dave