Dual band processing as OSC vs HOO

I'm curious as to how other people like to process their dual-band data on OSC cameras. I have about 20h of HaOiii on M101, my main goal was to do this to boost the Ha to add to upcoming broadband data.

In my OSC image (Stretching + green noise removal), it gives a very cold look to it, and the Ha layer has a bit of an orange/yellow hue to it. But the red is less pronounced, which defeats the goal of this specific shoot.

In the HOO image (Stretching), it looks warmer, as the Ha is more pronounced, and the Oiii is more of a gray color than a blue.

Images have had their resolution dropped by 66% for upload :o

Curious to hear your opinions!

HOO

OSC

Hi Kyle,

I'm on the newer side but thought I could share some of my previous experience with Dual Narrowband OSC images. I like to split the OSC image channels into R, G, B then select R -> Ha and the better of G/B for Oiii and work with them as monochrome. I then use as many copies of those for whichever composition I'm going for (HOO, SHO, Ha as L for SOHb, etc.). Splitting the channels also has the added benefit of setting you up for the broadband addition where you add the Ha to red and Oiii to blue and/or green.

If you wanted to be more accurate, you can use PixelMath or its equivalents (I recently found out Affinity Photo's is Procedure Texture) to generate an Oiii and a Ha image from the OSC image using some formulas for channel contributions based on your camera. However, keep in mind that since our OSC cameras have overlapping responses for each channel, we will always have some signal mixing in each color channel.

Clear Skies!
Joel

I have a mono camera. I made SHO image with LRGB stars, traditionnal way. I am not a specialist, but for me, my last improvements are:
- boost with specific formula the Oiii & Sii layers before mixing
- import image without stars in photoshop (save as tif file, 16 bits) in order to improve the colors.
- separate stars & the rest of the target in order to work on the two item differently in Pixinsight

Hello Kyle,

Whenever I shoot galaxies, I always shoot them in broadband/rgb color for the normal look and appearance. I then shoot the NB data and extract Ha and use that channel to clean up the “coldness” you’re referring to on the galaxy and also highlight the Ha spots in the galaxy. Here is my example of the same image you have taken here https://www.astrobin.com/p1eu5z/L/

A galaxy is more a broadband target unless you’re trying to do or pull something different from the image. If I were you I’d shoot this now in RGB broadband and because you have a good amount of data in Ha, I’d add the Ha to the broadband image. This would help pull out the Ha pockets and warm up your image.

Just my take on shooting this and most other galaxies.

Dale

Capturing dual band (at least with a L-Ultimate) is essentially an HOO representation. So when I do a nebula I'll just use the dual band OSC as is (substituting rgb stars). Now, when I do a broadband target that has Ha and/or O3 I'll split the dual band into Ha and O3 and do the continuum subtraction and add back into the original broadband OSC.

If during pure narrowband, going with the original colours or HOO depends on how you want the image to look. Doing HOO will usually bring more noise since the Oiii needs to be boosted.

If you’re combining dualband date with pure RGB, extract the Ha (red) and Oiii (green since it’s teal and OSC has more G pixels) and combine them with the red and blue RGB channels.

If during pure narrowband, going with the original colours or HOO depends on how you want the image to look. Doing HOO will usually bring more noise since the Oiii needs to be boosted.

If you’re combining dualband date with pure RGB, extract the Ha (red) and Oiii (green since it’s teal and OSC has more G pixels) and combine them with the red and blue RGB channels.

What would be the best way to combine like-colored channels? (Say Ha and red).
Would you perform something like a 2 frame stack with the two, or would you combine them with Lighten or transparency in Photoshop and flatten it?

I'm by no means an expert, and have been experimenting with an OSC with the Antlia Dual Band filters (HaOiii, and HbSii). I did try the channel separation but tbh wasn't sure it added much. It was a lot of effort, and the image didn't really improve (in fact it would come out with very bizarre looking colour casts). Whereas if I just processed and integrated them as OSC lights, I think the colours and tone came out quite pleasantly and without any coldness. This is from Bortle 9: https://www.astrobin.com/full/1ucc2b/C/ (All c&c welcome btw)

Dale Penkala:
Hello Kyle,

Whenever I shoot galaxies, I always shoot them in broadband/rgb color for the normal look and appearance. I then shoot the NB data and extract Ha and use that channel to clean up the “coldness” you’re referring to on the galaxy and also highlight the Ha spots in the galaxy. Here is my example of the same image you have taken here https://www.astrobin.com/p1eu5z/L/

I agree with Dale. Galaxies should be imaged always as LRGB and then (depending on target) add the dual band data.
This is an example, easy to try (M33):
https://www.astrobin.com/7vbf8f/?nc=collection&nce=19244

Hello, this is irrelevant to the topic, I'm just passing by, but something caught my attention:

Your OSC stars look almost exactly like mine (I'm a newt user), and I have noticed that my stars are too pale in color, usually being mostly brown or white.

Do you know why this is? I'm hoping that all those nice colors, like the strong blues and reds are able to come out, and that I'm just lacking integration time.

Expecting colors more like this: https://www.astrobin.com/qriud9/

Hello, this is irrelevant to the topic, I'm just passing by, but something caught my attention:

Your OSC stars look almost exactly like mine (I'm a newt user), and I have noticed that my stars are too pale in color, usually being mostly brown or white.

Do you know why this is? I'm hoping that all those nice colors, like the strong blues and reds are able to come out, and that I'm just lacking integration time.

Expecting colors more like this: https://www.astrobin.com/qriud9/

So in the in the images I posted, the star colors are going to look weird due to being only narrowband data (red and blue). At the end of the project, my starmask will just be data from the RGB data to get correct colors.
As for the color intensity, the true colors are usually muted, and will need a bit of a saturation boost to see the difference. In most of my images I try to leave them more 'natural' looking, but that's all personal preference.

Processing Method?

Related discussions