Some questions from a Mono n00b about process, starnet, RGB v Pixelmath, linear fit etc

[900SL]

Hello Peers

I've just embarked in the mono world and have some questions about HOO process

I shot 2 hrs Ha and 2 hrs Oiii the other night on the Heart.

B5, target near zenith, new moon. 533MM Pro

I calibrated/registered/stacked the Ha and Oiii separately in Siril. The Ha is pretty impressive when autostretched (coming from an OSC background). The Oiii is pretty noisy as expected. I then registered these two 'master' linear frames, as the Oiii was slightly out of alignment

So now I have two registered aligned linear images, one Ha and one Oiii, no other process applied: Compressed JPEGs: 

 

I am assuming the following steps in processing:

1: DBE each channel (Siril)

2: Deconvolution each channel (Siril)? Seems to work well on Ha, no good on Oiii using default settings

3: Denoise each channel? This is OK on Ha but causes/reveals some artifacts on Oiii using Siril defaults, given how grainy the Oiii is

4: Linear match? Not used this before but found it in Siril, I assume it is used on linear data to match different histograms? Which channel should be the reference?

I'm then at a crossroads, as I'm not sure how far you go with linear and when to use Starnet 2, Pixelmath, or RGB recomposition. I found Starnet 2 in Siril to work best on Linear data (Starnet 2 applies a default stretch to allow star removal)

(a) I could recombine the linear Ha and Oiii in PixelMath, then Starnet separation to work on the stretching. This seems to result in a lot of noise swamping the clean Ha in the final image.

(b) Or I could Starnet the Ha and Oiii separately, then stretch each channel. Again I get a load of noise in the Oiii which is very noticable in the recombination

 

I would welcome any tips on the best workflow for Mono HOO processing. I also recognise I need a lot more Oiii

 

 

 

 

[900SL]

This is a first stab at it, I used RGB composition after linear matching Oiii to Ha, which seemed to result in a less noisy image (at least I think that was the way I did it). Ha Red, Oiii green and blue

How on earth do you bring out the Oiii teal blue?

[WolfieGlos]

I’m intrigued by this, where we’ve had the discussion elsewhere about mono or OSC.

Sorry I can’t help on the questions, but Siril’s tutorials page does have quite an in-depth run through using mono data and using RGB Composition, Linear Fit and Pixel Math ; https://siril.org/tutorials/rgb_composition/

It appears to suggest to Linear Fit Ha to the OIII.

Hope it’s of help.

[Elp]

HOO gives a colour palette similar to RGB OSC images, if you want the blue to come out you have to replicate the SHO, as you dont have S2 it looks like you'll have to map as HHO to R, G and B respectively, green noise removal will help reduce the green mapped hydrogen (or you can tone down the green curve) and provides more of the golden yellow/blue pallette, but note due to one less channel of data it won't look exactly like an SHO mono image and some colour alteration will definitely need to be done.

Edited January 15 by Elp

[900SL]

Thanks for the feedback Gents. Reading up on it, the Heart doesn't have a lot of Oiii and I would need to double the Oiii integration as a minimum. 

I have an Sii filter on order, and I found a useful video by Cuiv on bringing up blue 💙,  which uses 'range masks'. I don't have either PS or Pixinsight, so that's not an option, unless Affinity Photo now does this.

Lots of fun regardless, I'm enjoying this new chapter 

[Elp]

GIMP works similarly to PS and it's free. I much rather use DTP software than something like PI to post process as you have more flexibility to do what you want.

Siril I only use to stack and pre process, all the colour adjustments and layer combinations I do with DTP.

[WolfieGlos]

On 13/01/2024 at 10:56, 900SL said:

I've just embarked in the mono world and have some questions about HOO process

I shot 2 hrs Ha and 2 hrs Oiii the other night on the Heart.

Not sure if this is the right topic to post this in, but with mono, are you taking flats for each filter? If so, assume you have to do this with a flat panel (outside) before changing filters?

[900SL]

55 minutes ago, WolfieGlos said:

Not sure if this is the right topic to post this in, but with mono, are you taking flats for each filter? If so, assume you have to do this with a flat panel (outside) before changing filters?

Yes, but I did them at the end, which might not be the right way thinking about it.

This mono lark is quite something. 1 hours Ha subs on HH at 25 degree altitude! And no calibration frames at all yet

 

[WolfieGlos]

Thanks - and wow, that's seriously impressive for only 1 hour! Incredible.

The flats side of it got me thinking about what's the best way to do it with all of the filters and the varying dust on each one. A DSD is another expense for me to consider if that's the proper way to do it.