Proccesing images from two setups OSC (1 st cam) + L(2 nd cam) it's possible and how ?

A dual setup is always great, especially when acquiring 2 objects at the same time. APP is made to just work for combining different setups even when the FOV is quite different. But if you're close that's probably better, you can integrate the data from the 2 setups like you normally do and then combine those results again with each other.

Posted by: @vincent-mod

A dual setup is always great, especially when acquiring 2 objects at the same time. APP is made to just work for combining different setups even when the FOV is quite different. But if you're close that's probably better, you can integrate the data from the 2 setups like you normally do and then combine those results again with each other.

How exactly do you do the bolded part? I'd like to do this, and I think I know what I want to happen, but I'm not positive or sure it's the correct way to do it. I would like to use the captured luminance data in the RGB frames in combination with the mono-L data.

I think what I want to do is to split my OSC data into a synthetic L channel + RGB at the individual sub level, then integrate the mono-L subs with the synthetic-L subs. Integrate the RGB subs as normal, then use the LRGB combine tool at the end. However, it appears that this workflow is not possible, the "Split channels" tool in the Calibrate tab only spits out RGB, which makes me think this is not the intended workflow.

Does the LRGB combine tool automatically create a synthetic-L channel from the RGB data, and add it to the provided L data?

For regular data, you can just load those final integrations in again and then register, normalize etc them like usual. You may want to manually set the reference frame in such a case.

Regarding your specific case, I wonder if combining synthetic luminance with actual luminance is a good idea. Real luminance data is likely better? I'll have to ask Mabula to be sure there. But yeah normally the split channels option is there for RGB data, to be able to combine that type of data with L, Ha, OIII etc.

For what you want here is a small explanation from Mabula (pretty old already but still applies I think):

First of all, you can easily create a synthetic lum of your data, there are several ways to accomplish this:

1. register, normalize and integrate all subs of all channels into 1 integration (tab 6, top option)
2. use the RGB combine tool to create a Luminance layer from the separate channels by giving the channels separate luminance weights and different multipliers. Usually, you need to multiply H-alpha very strongly compared to the broadband channels. A multiplier of 5 is very normal.

Save this luminance to a separate file.

Then create a new composite, like an LHaRGB, and use the luminance channel that you just created

The way I have done it is to put all of the data from one rig in session 1 and all the data from  the other rig in session 2.

I have a permanent set up so unless I alter things or ther has been a problem the cali frames are from the library.

Posted by: @vincent-mod

For regular data, you can just load those final integrations in again and then register, normalize etc them like usual. You may want to manually set the reference frame in such a case.

Regarding your specific case, I wonder if combining synthetic luminance with actual luminance is a good idea. Real luminance data is likely better? I'll have to ask Mabula to be sure there. But yeah normally the split channels option is there for RGB data, to be able to combine that type of data with L, Ha, OIII etc.

For what you want here is a small explanation from Mabula (pretty old already but still applies I think):

First of all, you can easily create a synthetic lum of your data, there are several ways to accomplish this:

1. register, normalize and integrate all subs of all channels into 1 integration (tab 6, top option)
2. use the RGB combine tool to create a Luminance layer from the separate channels by giving the channels separate luminance weights and different multipliers. Usually, you need to multiply H-alpha very strongly compared to the broadband channels. A multiplier of 5 is very normal.

Save this luminance to a separate file.

Then create a new composite, like an LHaRGB, and use the luminance channel that you just created

I got some L data and tried this, it worked pretty well. Finding a good weighting factor is tricky, I ended up using the noise value. I tried SNR, quality, noise, and equal, noise ended up looking the nicest. I had some pretty heavy walking noise in one of the OSC datasets, which may have thrown off the SNR calculation, which was my first try. I suspect, but am not positive, that integrating both datasets at the same time would make it easier or maybe less subjective for weighting based on quality/SNR/noise/etc.

In my case I have an APS-C Mirrorless camera (Fuji), and an asi183mm all on the same telescope. The mono data is much better on a per-time basis, but being able to combine them is nice, I get the big FoV of the APS-C and can use mono data to clean up areas as I deem necessary. This is an awesome capability to have.

Nice! So I talked with Mabula and he mentions that it kind of depends when to mix synthetic with real luminance. But it can certainly help as long as you see a noise improvement, it should be fine to do actually.