[RCC] Soul Nebula: Where is my OIII? Requests for constructive critique · FilippoTib · ... · 13 · 602 · 5

Miguel T.:
There is FAR more hydrogen than Oiii/Sii on most nebulas, and by a ridiculous ratio. This is where mono cameras shine because they can isolate the signal of individual atoms and give each of them the attention/integration time they need. I'm assuming you'll have to play around with PixelMath and make a best effort extracting your Oiii from the mixed signal, then stretch it independently before recombining. You'll most likely always end up with either not enough integration time for the Oiii or have your red pixels waste precious clear skies gathering hydrogen that you already have enough of.

For your second point, I can see a bit of the orange peel effect when zooming in on your image, which is usually caused by denoising a too noisy image. Even more integration time would help. For example, when I did this nebula, I went with 6 hours of pure Ha, 18 hours of pure Oiii and 8 hours of pure Sii, and that was with 3nm filters.

One thing you can try that work well on some nebulas. Since hydrogen is almost present everywhere, if you have enough Ha signal that only a small amount of denoise is required and that you manage to extract it, use it as luminance layer with LRGBCombination.

Miguel T.:
There is FAR more hydrogen than Oiii/Sii on most nebulas, and by a ridiculous ratio. This is where mono cameras shine because they can isolate the signal of individual atoms and give each of them the attention/integration time they need. I'm assuming you'll have to play around with PixelMath and make a best effort extracting your Oiii from the mixed signal, then stretch it independently before recombining. You'll most likely always end up with either not enough integration time for the Oiii or have your red pixels waste precious clear skies gathering hydrogen that you already have enough of.

For your second point, I can see a bit of the orange peel effect when zooming in on your image, which is usually caused by denoising a too noisy image. Even more integration time would help. For example, when I did this nebula, I went with 6 hours of pure Ha, 18 hours of pure Oiii and 8 hours of pure Sii, and that was with 3nm filters.

One thing you can try that work well on some nebulas. Since hydrogen is almost present everywhere, if you have enough Ha signal that only a small amount of denoise is required and that you manage to extract it, use it as luminance layer with LRGBCombination.

Hello. Thank you very much for the feedback!

Yes, I was aware of the hydrogen imbalance. However, I did not think so much! Also interesting about the separate stretch. I had always considered linear phase recombination. I'll have to try the technique you recommended! Thank you!

I will try integrating more frames and noise reduction then. 

Thank you so much again for all the advice and good skies!
F.T.

Miguel T.:

Miguel T.:
There is FAR more hydrogen than Oiii/Sii on most nebulas, and by a ridiculous ratio. This is where mono cameras shine because they can isolate the signal of individual atoms and give each of them the attention/integration time they need. I'm assuming you'll have to play around with PixelMath and make a best effort extracting your Oiii from the mixed signal, then stretch it independently before recombining. You'll most likely always end up with either not enough integration time for the Oiii or have your red pixels waste precious clear skies gathering hydrogen that you already have enough of.

For your second point, I can see a bit of the orange peel effect when zooming in on your image, which is usually caused by denoising a too noisy image. Even more integration time would help. For example, when I did this nebula, I went with 6 hours of pure Ha, 18 hours of pure Oiii and 8 hours of pure Sii, and that was with 3nm filters.

One thing you can try that work well on some nebulas. Since hydrogen is almost present everywhere, if you have enough Ha signal that only a small amount of denoise is required and that you manage to extract it, use it as luminance layer with LRGBCombination.

Hello. Thank you very much for the feedback!

Yes, I was aware of the hydrogen imbalance. However, I did not think so much! Also interesting about the separate stretch. I had always considered linear phase recombination. I'll have to try the technique you recommended! Thank you!

I will try integrating more frames and noise reduction then. 

Thank you so much again for all the advice and good skies!
F.T.

The more you can get away with the least noise reduction, the better it is. Staying at under 70% with NX is preferable. Also, a big game saver is using the Ha as luminance as I said. There is just so much Ha signal that it usually only need a slight denoise, even at low integration times. Using this layer as luminance help to mitigate the noise of Oiii/Sii and you can play around with the Chrominance noise reduction of LRGBCombination as well. Also, don't hesitate to go well above your 12 hours integration time. At this time of the year it was only a bit more than one full night. 2 more nights and you're already at 30 hours.

Finally, here's an image I posted in the description of my latest image of the Heart nebula, giving a preview of how each individual filters look. Each filter had ~14 hours of data and you can clearly see how noisy the Oiii is compared to the others. I believe I had 65% denoise on Oiii and 40% on Ha:

Miguel T.:
Using this layer as luminance

Miguel T.:
There is FAR more hydrogen than Oiii/Sii on most nebulas, and by a ridiculous ratio. This is where mono cameras shine because they can isolate the signal of individual atoms and give each of them the attention/integration time they need. I'm assuming you'll have to play around with PixelMath and make a best effort extracting your Oiii from the mixed signal, then stretch it independently before recombining. You'll most likely always end up with either not enough integration time for the Oiii or have your red pixels waste precious clear skies gathering hydrogen that you already have enough of.

For your second point, I can see a bit of the orange peel effect when zooming in on your image, which is usually caused by denoising a too noisy image. Even more integration time would help. For example, when I did this nebula, I went with 6 hours of pure Ha, 18 hours of pure Oiii and 8 hours of pure Sii, and that was with 3nm filters.

One thing you can try that work well on some nebulas. Since hydrogen is almost present everywhere, if you have enough Ha signal that only a small amount of denoise is required and that you manage to extract it, use it as luminance layer with LRGBCombination.

Hello. Thank you very much for the feedback!

Yes, I was aware of the hydrogen imbalance. However, I did not think so much! Also interesting about the separate stretch. I had always considered linear phase recombination. I'll have to try the technique you recommended! Thank you!

I will try integrating more frames and noise reduction then. 

Thank you so much again for all the advice and good skies!

Dominik:

Miguel T.:
There is FAR more hydrogen than Oiii/Sii on most nebulas, and by a ridiculous ratio. This is where mono cameras shine because they can isolate the signal of individual atoms and give each of them the attention/integration time they need. I'm assuming you'll have to play around with PixelMath and make a best effort extracting your Oiii from the mixed signal, then stretch it independently before recombining. You'll most likely always end up with either not enough integration time for the Oiii or have your red pixels waste precious clear skies gathering hydrogen that you already have enough of.

For your second point, I can see a bit of the orange peel effect when zooming in on your image, which is usually caused by denoising a too noisy image. Even more integration time would help. For example, when I did this nebula, I went with 6 hours of pure Ha, 18 hours of pure Oiii and 8 hours of pure Sii, and that was with 3nm filters.

One thing you can try that work well on some nebulas. Since hydrogen is almost present everywhere, if you have enough Ha signal that only a small amount of denoise is required and that you manage to extract it, use it as luminance layer with LRGBCombination.

Hello. Thank you very much for the feedback!

Yes, I was aware of the hydrogen imbalance. However, I did not think so much! Also interesting about the separate stretch. I had always considered linear phase recombination. I'll have to try the technique you recommended! Thank you!

I will try integrating more frames and noise reduction then. 

Thank you so much again for all the advice and good skies!

Hello,

I'm using an ASI ZWO 533MC Pro with a DuoNarrowband filter (Askar CM D1). Just keep in mind that the OIII portion of an OSC camera with a Bayer matrix is only 25% and is often heavily overlaid by strong HA. This means that compared to mono, you'll need to invest more integration time and sometimes you have to separate the two channels during image processing.

Of course, you can use scripts like Bill Blanchet's, but I find that working on separate Channels gives you a bit more flexibility.  I use various techniques; sometimes I even blend HA (red) and OIII (blue) together and when it suits the object, i also use the Narrowband Normalization Script.

I find the tip about using HA as luminance quite interesting. I've noticed that with longer integration times, OIII can become quite noisy. All in all, i'm considering switching to mono because of the greater flexibility in image processing and simply because you get more signal per channel.

By the way, I've found the scripts Selective Color Correction, Image Blend, and Color Mask Mod to be very helpful with color adjustments.

CS
Dominik