Synthetics with OSC

User images created with StarTools.
Post Reply
kongpup
Posts: 4
Joined: Sun Oct 02, 2022 10:29 pm

Synthetics with OSC

Post by kongpup »

Here's one of those situations where my knowledge of a subject is so limited that I don't even know what question to ask, so I'm including these images of M27 instead.
Upper two images (left w/ L Enhance filter, right w/ Antlia ALP-T). These were both processed using the Star Tools recommended method of layering a dualband image with an image made with an IR cut filter (Baader IR cut in both). After using ST for a couple of years, I only recently bothered to actually read the directions for the Color module. Previously, I had used only a dualband image, which frequently would come out with dayglow red highlights, requiring further processing to tone down. The 2 image procedure made quite a difference.
Lower two images (same order). Had recently become interested in what I can only describe as "synthetic" SHO & HOO images with an OSC. I used the Compose mode for these, with "BiColor from OSC" chosen in "Luminance, Color" field, Luminance file left empty, and the same Deep Sky Stacker Autosave file in each of the RGB fields. In the Color module, the "Bicolor" option was chosen, and in the "Matrix" field, I honestly just kept pushing different options until hitting on one that looked like an "official" looking image found on the web.
Can anyone tell me what I've come up with here - if it was done correctly or not or any other comments?
Attachments
M27 filter compare.jpg
M27 filter compare.jpg (233.39 KiB) Viewed 2079 times
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Synthetics with OSC

Post by admin »

kongpup wrote: Mon Oct 03, 2022 1:48 am Here's one of those situations where my knowledge of a subject is so limited that I don't even know what question to ask, so I'm including these images of M27 instead.
Upper two images (left w/ L Enhance filter, right w/ Antlia ALP-T). These were both processed using the Star Tools recommended method of layering a dualband image with an image made with an IR cut filter (Baader IR cut in both). After using ST for a couple of years, I only recently bothered to actually read the directions for the Color module. Previously, I had used only a dualband image, which frequently would come out with dayglow red highlights, requiring further processing to tone down. The 2 image procedure made quite a difference.
Lower two images (same order). Had recently become interested in what I can only describe as "synthetic" SHO & HOO images with an OSC. I used the Compose mode for these, with "BiColor from OSC" chosen in "Luminance, Color" field, Luminance file left empty, and the same Deep Sky Stacker Autosave file in each of the RGB fields. In the Color module, the "Bicolor" option was chosen, and in the "Matrix" field, I honestly just kept pushing different options until hitting on one that looked like an "official" looking image found on the web.
Can anyone tell me what I've come up with here - if it was done correctly or not or any other comments?
Hi,

It's indeed a little tricky to answer a question that is not terribly well-defined. I suppose your question is "what do colours in narrowband mean"?

In order to answer that you'd first like to understand the following;
  • Are you across where different emission bands sit in the spectrum?
  • Are you across what Luminance, SHO, HOO, etc. refer to?
  • Do you know which emission bands are not recorded by your gear/filters (or which bands are actively filtered out by your filters, etc.)?
and finally;
  • What is you reason and goal for acquiring narrowband?
Is that helpful in breaking down your question?
Ivo Jager
StarTools creator and astronomy enthusiast
kongpup
Posts: 4
Joined: Sun Oct 02, 2022 10:29 pm

Re: Synthetics with OSC

Post by kongpup »

admin wrote: Mon Oct 03, 2022 10:27 pm Hi,
It's indeed a little tricky to answer a question that is not terribly well-defined. I suppose your question is "what do colours in narrowband mean"?

In order to answer that you'd first like to understand the following;
  • Are you across where different emission bands sit in the spectrum?
  • Are you across what Luminance, SHO, HOO, etc. refer to?
  • Do you know which emission bands are not recorded by your gear/filters (or which bands are actively filtered out by your filters, etc.)?
and finally;
  • What is you reason and goal for acquiring narrowband?
Is that helpful in breaking down your question?
I have a few years of experience with the dualband filters and have a fair amount of knowledge of their application. What is new to me is the so-called synthetic SHO & HOO imaging with an OSC (mine being the ZWO294mcp). My questions then would be, 1) would the lower images be considered correctly done synthetics and, 2) is my described procedure to process them correct? (my posted description didn't include all the other standard steps, Ie. AutoDev, Wipe, Decon, etc. all of which were also done).
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Synthetics with OSC

Post by admin »

kongpup wrote: Wed Oct 05, 2022 1:02 am What is new to me is the so-called synthetic SHO & HOO imaging with an OSC (mine being the ZWO294mcp). My questions then would be, 1) would the lower images be considered correctly done synthetics and, 2) is my described procedure to process them correct? (my posted description didn't include all the other standard steps, Ie. AutoDev, Wipe, Decon, etc. all of which were also done).
Thank you, though I am still a little confused as to what you mean by synthetics/synthetic in this context...

Creating SHO composites from a duo/tri/quadband filter is not possible, due to S-II and H-alpha occupying the same wideband red channel. Depending on whether your filter isolates O-III or lobs it in with H-beta in the green and blue channels, O-III may be less precise (and may correlate substantially with Ha).

If your filter, however, isolates Ha (red channel) and O-III (blue and green), you can image S-II separately and then combine this into a proper SHO dataset. You do not mention an S-II filter, so I don't think this is what you are doing, or what you are after? (hence my questions as to what your goals are, and whether you are across what things like "SHO" refer to etc.)

For the record, your L-Enhance filter isolates Ha in the red channel, but lobs in H-beta with O-III in the green and blue channels;
Image
The Antlia filter is a true H-alpha / O-III filter, which is ideal if you are looking to do SHO composites (by separately acquiring data through an S-II filter);
Image
I would not also use the IR-cut filter when using either filters, because they already cut IR/UV, given that they only pass Ha and O-III/Hb. (hence my question as to whether you are across what parts of the spectrum recorded/rejected by the filters)

With both filters, you can make bi-colors by mapping Ha to a color and mapping O-III (or O-III+Hb in the case of the L-Enhance) to another color. A bi-color that is closest to what these emissions would look like in the visual spectrum, would indeed be the tried and tested HOO mapping. This yields a red hue for Ha and a cyan hue for O-III (or O-III+Hb in the case of the L-Enhance).

If the above is your goal, then the procedures you describe (e.g. they settings and way of importing the dataset in the Compose module, and the bi-color preset selection in the Color module) sound correct. They would indeed yield the upper two images. The bottom two color mappings appear to map Ha to green and O-III to blue. While absolutely still useful to distinguish the two emissions it is a less common bi-color. If this is the result of choosing a SHO mapping, then S-II (variations in the red channel when mapped as RGB:SHO) is, of course, not going to magically show up; it was never recorded.

Does that help?
Ivo Jager
StarTools creator and astronomy enthusiast
Mike in Rancho
Posts: 1141
Joined: Sun Jun 20, 2021 10:05 pm
Location: Alta Loma, CA

Re: Synthetics with OSC

Post by Mike in Rancho »

Hi Kongpup, and welcome!

I too am somewhat (okay, perhaps fully) confused by what you are asking, or what you did. :confusion-shrug:

By "synthetic" do you mean the color mapping, as in is it "faux"? Or do you mean something regarding the Synthetic luminance from compose?

As to the two image layering, do you mean that you added RGB stars to a narrowband image? That's moderately common and there is a logic to it, but I'm not sure I'm actually seeing that in your images.

If, however, for your bicolor layer, you composed as described into bicolor from OSC, hit the bicolor preset in color, and then simply chose from among the various matrix options, those should still be legit bicolors, even if you ended up using a tricolor SHO matrix. It would not turn things into tricolor, however. As part of the composite you selected, the G and B channels are combined into one, and then that identical file ends up back in the G and B.

Thus, any mapping that you choose, including SHO, still ends up creating only two hues - one for Ha, and one for OIII. But them hues are a little different than the hues you might get from one of the shorter list of bicolor mappings. I traced out what a few of those SHO mappings do on a notepad, when applied to a bicolor composite.

I believe Martin (Startrek) employs this frequently with his bicolors. If I am understanding it right? :?:
Startrek
Posts: 351
Joined: Mon Dec 30, 2019 3:49 am

Re: Synthetics with OSC

Post by Startrek »

Yes like Mike said I use a 2600MC with an L Extreme filter ( now upgraded to the Antlia ALP T ) and produce Bi Colour images via the Compose module ( OSC Bi Color )

I thought along the same lines as you in regards to what colour is the correct representation

With OSC and dual band filters due to the absence of Sii colours tend to be quite unusual on certain emission nebula even using the Matrix channel blends creates some weird and funky colour blends

On some Emission nebula like the Lobster , Cats Paw etc… I can’t even produce a Bi colour rendition. The Oiii and Sii signals must be very weak and a OSC struggles with NB signal in any case due to the Bayer Matrix

Some folk use OSC with their Dual band filters , capture Ha and Oiii and then use a separate Sii filter and in doing so attempt to process true narrowband images. Some are quite successful, others struggle

I’m moving to Mono next year (2600MM ) with quality LRGB and Narrowband filters ( Chroma ) as I’ve gone as far as I can with OSC bi color after 3 years at it. It does OK but does have its limitations.

I image under both sides of the spectrum , Bortle 8 City suburban in my heavy light polluted backyard and Bortle 3 skies at my rural coastal retreat which has a NexDome

Another learning curve with Mono , but excited to move into higher end imaging

Clear Skies

Martin
kongpup
Posts: 4
Joined: Sun Oct 02, 2022 10:29 pm

Re: Synthetics with OSC

Post by kongpup »

admin wrote: Wed Oct 05, 2022 1:37 am Does that help?
Mike in Rancho wrote: Wed Oct 05, 2022 6:08 am By "synthetic" do you mean the color mapping, as in is it "faux"? Or do you mean something regarding the Synthetic luminance from compose?
I believe Martin (Startrek) employs this frequently with his bicolors. If I am understanding it right? :?:
Startrek wrote: Wed Oct 05, 2022 9:14 am Yes like Mike said I use a 2600MC with an L Extreme filter ( now upgraded to the Antlia ALP T ) and produce Bi Colour images via the Compose module ( OSC Bi Color )
I thought along the same lines as you in regards to what colour is the correct representation
With OSC and dual band filters due to the absence of Sii colours tend to be quite unusual on certain emission nebula even using the Matrix channel blends creates some weird and funky colour blends
Martin
Thank you, Ivo, Mike, and Martin
It was actually the back and forth between Mike & Martin (SHO with OSC? post) as well as a CN post about creating an HOO from an OSC that piqued my interest in this type of processing.
That said, i would like to put aside the 2 upper images (they were processed using my now "regular" ST procedure and I'm happy with them - they were included only for comparison.
I'm aware that I can't get tri color with a Dualband filter without adding Sii and am not trying to do so. In the SHO with OSC? post, Ivo outlined a procedure for processing a bicolor image that I've tried to follow after trying to fill in a few blanks in said procedure. Here was my workflow for the lower images (Dualband filters only on each, no IR cut):
Compose (BiColor from OSC preset - this is where the word "synthetic" popped up), Luminance, Color left empty, RGB files all with the same DSS Autosave file >
AutoDev>Bin>Crop>Wipe>AutoDev>Contrast>HDR>Sharp>Decon>
Color (BiColor preset, Matrix - one of the HO choices - I believe it was (H+O)HO>
Track/NR
That's it - as Martin said, many of the Matrix choices produced strange effects, the HO was the only one that produced something similar to other published images that I found. So all I'd like to know is if that workflow seems correct or if I've come up with something weird with the lower images.
Thanks again,
Chris
Mike in Rancho
Posts: 1141
Joined: Sun Jun 20, 2021 10:05 pm
Location: Alta Loma, CA

Re: Synthetics with OSC

Post by Mike in Rancho »

Hi Chris,

Hmm, I'm not sure what you mean by Luminance, Color left empty in Compose. Can it be? I don't think I've seen a blank option, and what would that be anyway? Do you mean the Luminance file slot was empty? If so, that is actually appropriate for your single duoband stack. The "synthetic" L, or usually "L + Synth L" (which I agree can be potentially confusing) is just the creation of the luminance layer that you will process your details in, until you get to Color. This is default and key to how much of ST works, parallel processing L and color. You can get the same with a center-option open of an OSC file, though not the "bicolor" that you'd want here.

I don't see anything untoward in your processing workflow. The chosen mapping would certainly be a legitimate bicolor. As long as one considers OSC duoband to be akin to narrowband false color. I certainly think it does. As such, you can map your hydrogen and your oxygen to whatever colors you find appropriate, or just prefer. A straight up HOO is the most common, which also puts the H in the red and the O in the blue or teal, and that's kind of generally where those wavelengths would fall anyway. But one doesn't have to do that. In SHO we make our H green. No reason it can't be green in a bicolor too! :D

Is that what you meant by seeing having seen such images of this target on the web? It sounds to me like they were tricolor SHO.

Your result is still spot on. You seem to have green for H and a purple for the O. You can play with the RGB bias sliders to throttle these emissions as needed to reveal what you would like to reveal, as described in the docs.

Perhaps moreso with other targets, another fully legit bicolor-only mapping that sort of often resembles common tricolor SHO is H(H+O)O.

Cool stuff!
kongpup
Posts: 4
Joined: Sun Oct 02, 2022 10:29 pm

Re: Synthetics with OSC

Post by kongpup »

Mike in Rancho wrote: Thu Oct 06, 2022 3:02 am Hmm, I'm not sure what you mean by Luminance, Color left empty in Compose. Can it be? I don't think I've seen a blank option, and what would that be anyway? Do you mean the Luminance file slot was empty? If so, that is actually appropriate for your single duoband stack. The "synthetic" L, or usually "L + Synth L" (which I agree can be potentially confusing) is just the creation of the luminance layer that you will process your details in, until you get to Color. This is default and key to how much of ST works, parallel processing L and color. You can get the same with a center-option open of an OSC file, though not the "bicolor" that you'd want here.

I don't see anything untoward in your processing workflow. The chosen mapping would certainly be a legitimate bicolor. As long as one considers OSC duoband to be akin to narrowband false color. I certainly think it does. As such, you can map your hydrogen and your oxygen to whatever colors you find appropriate, or just prefer. A straight up HOO is the most common, which also puts the H in the red and the O in the blue or teal, and that's kind of generally where those wavelengths would fall anyway. But one doesn't have to do that. In SHO we make our H green. No reason it can't be green in a bicolor too! :D

Is that what you meant by seeing having seen such images of this target on the web? It sounds to me like they were tricolor SHO.

Your result is still spot on. You seem to have green for H and a purple for the O. You can play with the RGB bias sliders to throttle these emissions as needed to reveal what you would like to reveal, as described in the docs.

Perhaps moreso with other targets, another fully legit bicolor-only mapping that sort of often resembles common tricolor SHO is H(H+O)O.

Cool stuff!
First off - apologies Mike. I've just now seen your response - I either didn't get a notification or missed it. Yes, I did mean the Luminance file slot left empty. Thanks much, your response does answer my vague question (they were HOO images that I had found to be similar on the web).
Chris
Post Reply