Sh2-308 Dolphin Nebula Processing

[Startrek]

I’m still gathering data ( Ha and Oiii ) on the Dolphin Nebula and would like some processing tips prior to completing the capture.
This object is primarily Oiii with Ha accents

Any suggestions on how would I load Compose to achieve a good narrowband composite ( remember there is no Sii ) that reflects this object.

Comments , advice most welcome

Clear Skies
Martin
PS : Camera 2600MM
Filters Ha and Oiii 3nm Antlia pro

Hopefully integration around 12 to 15 hours of data ( mostly Oiii )

[decay]

Hi Martin, at a first thought I would try to process like data from a dual narrow band filter and OSC.
So, setting “Luminance, Color” to “Bi-Color from OSC/DSLR” and loading the Ha file into the red slot and OIII data into the green and blue slots.

As you stressed “Ha accents” I wonder if you would like to try to use the NBA module? Hm, interesting. This would mean loading the Ha data into the NB Accents slot. And probably OIII data into all three RGB slots. But I'm not sure which "Luminance, Colour" setting to choose in this case.

I’m looking forward to see your Dolphin!

Best regards, Dietmar.

[Startrek]

Hi Dietmar,
Thanks for your suggestion

I just found Guys notes on Steve Cannistras HOO composites , so I’m not sure whether the OSC Bi Color option will work with just 2 individual channels (Ha and Oiii ) from Mono data sets to use for HOO
The Bi Color option uses all three channels loaded with the same data set from OSC taking into account the Bayer Matrix to create a Synthetic Lum based on the weighting sliders being similar on all channels
See below from ST web notes …
"L + Synthetic L from R(2xG)B, R(GB)(GB) (Bi-Color from OSC/DSLR)" creates a synthetic luminance dataset from Luminance, Red, Green and Blue, weighted according to the exposure times provided by the "Total Exposure" sliders and taking into account the presence of a Bayer matrix. The colour information will consists of red as imported, with an average of green+blue assigned to both the green and blue channels. This mode is very useful for creating bi-colours from duo/tri/quad band filtered datasets.

My first thoughts were to set the Luminance , Color scheme to L + Synthetic L from RGB, RGB ( Mike advised last year this is probably your safest option with Mono data sets whether Broadband or Narrowband)
Then load Ha into R , Oiii into G and Oiii into B. My data sets will not be equal as Oiii will be probably double to that of Ha and therefore the weighting will be adjusted accordingly
I will try Steve Cannistras method to create a HOO
I’ve had excellent results so far with SHO as it’s fairly straightforward but HOO is more tricky
This Dolphins Head nebula is so damn faint , my 5min Ha subs don’t show any signs of the object and the 5 min Oiii subs just barely show a faint silhouette of the nebula
Obviously Bortle 8 skies don’t help
I’m thinking 15 hours may not be enough ?
This nebula is gone by April so hopefully 2 or 3 more clear nights in March should allow me to complete it
Thanks again
Clear Skies
Martin

[Mike in Rancho]

Hi Martin!

Oh boy, when I first saw this post I assumed you were doing this at the rural site. I think it may be the wiser choice. I mean hopefully the hours stack up well and you get something strong enough, but...

Moving on...Mike advised what? Golly, that doesn't sound like me. Apologies if I messed up.

For a bicolor with discrete filters, Ivo has us covered. Look for a similar bicolor option, but the one that does NOT say it is for OSC. Now, here's the trick and hopefully I remember it right: Of course you want (GB)(GB) identical and locked. Normal L+Synth L from RGB, RGB may result in the same Synth L (if you zero one of the G/B sliders), but I think will act differently in Color, permitting you to alter channel balances possibly improperly for bicolor. I could be wrong there and I have to think about it a bit more deeply as to what happens when the pixels match, but you are allowed to change the relative bias. Possibly just a hue change.

In any event, again the trick is to either load your OIII only into one of the G or B slots, it doesn't matter which, OR, load OIII into both G and B but zero one of the sliders out. This prevents over-weighting of the single OIII when creating the Synth L. This is somewhat different than the OSC bicolor for duoband, where the OIII data falls across two different bayer filters and so you want both, but the G gets a bump in weighting (about 1.4x) due to better SNR. Of course that is based simply on the bayer pattern (2xG) rather than the sensor QE between B and G.

When you get to Color, you can pick the Bicolor preset and everything should act similarly to the OSC bicolor you are used to.

In 1.9, you can use the extra Compose module options that will pre-select that going into Color, but I think you have to choose the OSC duoband option since there isn't one (yet?) for discrete Ha OIII bicolor from mono. Off the top of my head that just sets things up automatically rather than having to hit bicolor once in Color, and also (again from memory) affect whether one gets green cap or purple cap to choose from.

I'd have to go play around with some data to help out my recollection here. There are lots of ways you can use Compose, Color, and even NB Accent (may require extra prep work if you want both channels) to change things up a bit, as long as one is mindful of what is happening as well as, of course, if you want to hold yourself to keeping relative emissions balances.

[Startrek]

Hi Mike,
Thanks
I must have missed the composite color scheme below for Bi Color ( Mono ) …..
"L + Synthetic L from RGB, R(GB)(GB) (Bi-Color)" creates a synthetic luminance dataset from Luminance, Red, Green and Blue, weighted according to the exposure times provided by the "Total Exposure" sliders. The colour information will consists of red as imported, with an average of green+blue assigned to both the green and blue channels. This mode is suitable for creating bi-colours from, for example, two narrowband filtered datasets.

I’ve always used L+ Synthetic L from RGB , RGB since upgrading to Mono in August 2023

Ok , let me see if I’ve got your procedure correct , so select L + Synthetic L from RGB, R ( GB) ( GB) Bi Color
Load Ha into R ( Sii ) slot , load Oiii into G (Ha )slot , load Oiii again into B ( Oiii ) slot but move this slider back to zero !
Question: For Ha R slot and Oiii G slot should I adjust weightings according to exposure times of my captured data or leave as default ??



Also is it worth using Guys notes on Mono Bi Color procedures ( Steve Cannistra method ) ??

Note I haven’t upgraded to 1.9 yet still using 1.7 and 1.8

Martin

[Mike in Rancho]

Ok , let me see if I’ve got your procedure correct , so select L + Synthetic L from RGB, R ( GB) ( GB) Bi Color
Load Ha into R ( Sii ) slot , load Oiii into G (Ha )slot , load Oiii again into B ( Oiii ) slot but move this slider back to zero !
Question: For Ha R slot and Oiii G slot should I adjust weightings according to exposure times of my captured data or leave as default ??

Hi Martin,

Yes that's one of the viable ways to compose the mono bicolor. Or, again, just load Oiii once into either B or G, doesn't matter which.

For weighting, hard to say. Play around with it!

It all depends on how the data turns out. Ivo says it's all about SNR, right? But with NB, the exposure times may not be a perfect stand-in for resulting SNR of each filter channel. Oiii and Sii are usually far more noisy per equal exposure time than Ha. So sometimes I adjust accordingly. But you don't want to completely lower things too much, in case there's discrete Oiii or Sii data apart from Ha. Can't colorize what's not illuminated.

This is an oddball target though, so who knows? Kind of a cross between a dolphin-shaped Veil, and maybe the Crescent. Big and blue with some red internals.

Also is it worth using Guys notes on Mono Bi Color procedures ( Steve Cannistra method ) ??

Okay I checked this out, and am experiencing severe bafflement.

First, the Cannistra link describes something he is doing with PS layers, if I have that right. I don't have the motivation to sketch out his steps to see what the result is or if it makes any sense. Sorry! And I'm not sure the ST implementation provided is the same thing.

The first two steps, creating a newG out of Ha+OIII, and then recompositing that with Ha and OIII, seems to me a complete wash as far as the Synth L, regardless of the weighting. So why even do it? Somebody fact-check me in case I am off base in the compositing math here.

Then, when you get to Color, well all you've done is set things up to be akin to H(H+O)O, except lousy (as far as channel bias adjustments go), and we already have H(H+O)O anyway, which works as intended in bicolor mode.

Honestly I would avoid this like the plague.

Maybe we should even lobby for this to be excised from the User Notes. What am I missing here?

[Startrek]

Hi Mike,
Thanks
We shall see how this one pans out , yes this object is certainly challenging under less than ideal skies ( 27 of my Oiii 5min subs were under a near full moon as well )

I’ll ignore Guys notes relating to Bi Color and the Steve Cannistra’s link as I’m baffled as well

I’ll play around with weighting on the Ha and Oiii and see how it processes out.

Thanks again
Martin