Need advice on combining data

[ramv]

Hi,
I shot several nights of data worth on a single target for the first time. (actually just got in after driving over 2 hours from my dark site) Here are the characteristics of the capture

1. I captured Ha, L, R, G, B on both nights
2. On night 1, Ha was not binned, on night 2 I binned Ha because the channel was quite weak and I needed more signal. All Ha captures were 15 mins long.
3. L was captured for 5 min subs on night 1 and 10 min subs on night 2. Again the reason was to get a bit more signal.
4. R, G, B was binned 2x2 on both nights. All were 5 min subs
5. I shot the object with a slightly different framing on night 2. I wish to do a mosaic merge so that the merged view encompasses both the fields. Both of these fields have significant overlap.

Not sure if this info is relevant, but I shot at F5 and 530mm.

I would like to know which is the best way to proceed.

One possible approach I was considering was to process each night's data separately/do Ha blend and then do a mosaic merge. Another option is to combine all the individual channels for each night and then do a final channel combination + Ha addition.
Which of the above is the preferred approach to maximise s/n as it is quite a faint object? I also want to make sure that there are no merge lines visible if I went with approach 1 as a result of different sub lengths.

Is there any issue with Ha combination when some of the data is binned 2x2 and some of it is not?

Advice from folks who've done something like this in the past would be invaluable.

Best regards,
--Ram
 

[sreilly]

If I'm following you correctly it sounds as if you have two different images. Night one is centered differently than night two. Process each separately and then make your two panel mosaic. If you do have images from the same night at different binning then I would combine each separately and then combine after using the Dynamic align tool to align both images.

-Steve

[ramv]

Steve, thank you.
Does the mosaic combination process improve SNR in the overlapping areas? (similar to say stacking two images where overall SNR goes up almost 50%)

Thanks,
--Ram

[georg.viehoever]

Does the mosaic combination process improve SNR in the overlapping areas? (similar to say stacking two images where overall SNR goes up almost 50%)

If you use the Average combination mode in GradientMergeMosaic, the SNR in the overlap regions is reduced just as a combination with a mean operator would do it. However, this noise reduction compared to the neighboring regions may be visible to a viewer. If this is the case, you should use Overlay mode.

GradientMergeMosaic can only level intensity differences between regions. It cannot equalize patterned or unpatterned noise, or differences in resolution(general color/imaging artefacts.... These may still be visible after merging images. You therefore should be careful to make you 2 panels as similar as possible to each other before merging them.

Georg