Eta Carina Mosaic Experiment

[AstroViking (Steve)]

Hi all,

One feature of Kstars that I've been curious to try out is the 'Mosaic' feature, buried in the 'Sequencer' page. I watched a few YouTube videos so that I got an idea of how it worked and then decided to give it a try on the next clear night.

The selected Target? The Eta Carina nebula, of course. It's big, bright, and easy to find in the night sky.

I elected to use the 'stock' 420mm focal length of my telescope because I already had it configured within Ekos. The downside of this is the reduced field of view and the slightly elongated stars in the image corners. For the next mosaic I will prepare a bit earlier and use the flattener / reducer in the setup, and maybe drop the UV-IR cut filter in there as well. We'll see.

Also, the default 10% overlap is not enough - the stacking artefacts eat into this overlap and may make if difficult to align the panels. So I'll bump it up to 15%, or maybe even 20%, depending on what sort of FoV I have.

Because this was an experiment, I only took 90 frames for each panel - so about 22.5 minutes of integration time each. (The clouds were forecast to roll in at about the time I was going to finish. They didn't, but that's pretty usual for my luck when it comes to astro and the weather.)

I elected to do the stacking with SiriL, because it's the fastest and has very handy (and powerful) pre-processing tools. Each panel was stacked in SiriL, after which came a Background Extraction, Green Noise removal and a Histogram Stretch. Oddly, SiriL 1.0.6 has started crashing when I right-click the Background Extraction window to remove a sampling point. Not good, but given that SiriL 1.2.0 has entered 'beta' stage, I'm not overly worried.

The four stacked and pre-processed panels were loaded into Affinity Photo's Panorama tool and automatically stitched together. After cropping, I began the work of converting the very "dusty brown" looking output (see image V.1 below) into something that looked like an OSC image.

Image V.2 is a "mid-way" image, whilst V.3 is the final output. The colours are a bit over-saturated, I think, but are much closer to what my other images of Eta Carina look like.

For what it's worth, my working images were in the order of 70 megapixels (roughly 10,000 x 7000 pixels before cropping) and just under 900 megabytes in size.

As always, comments, feedback and constructive criticism are welcome!

Cheers,
V

Hardware: SW72ED @420mm / HEQ5-Pro / SV305+SV165 / ASI183MC-Pro / Kstars+Ekos

Image details:

* Gain: 50
* Offset: 10
* Temperature: 0 degrees C
* Exposure: 15 seconds
* Lights: 90 (22.5 minutes of data per panel)
* Darks: 50
* Flats: 50
* Dark Flats (Bias): 50
* Stacked and pre-proc'd in SiriL, then processed in Affinity Photo

[oska (John)]

Nice one V, that's a great result for only 20 mins. And a 2x2 experiment!? Way cool!
Have you looked at freebies like telescopius to post the full size?
How are you calculating exposure time?
The internal (eternal?) struggle with saturation seems to be a rite of passage? Quite often I'm finding my final tweak is to desaturate, sometimes quite a lot. It was definitely a surprise to me that I've had to fight it.

[AstroViking (Steve)]

Hi John,

Thanks. Now that I (roughly) know what I'm doing, the next attempt will be a bit more serious.

I used Telescopious to get an idea of how the image would turn out, but haven't looked at posting any images there. I haven't considered posting my work anywhere (other than here and on my website) because I don't think it's good enough.

When it comes to exposure times I take a series of frames, usually from 10 seconds up to about 60 and then eye-ball which exposure looks the best and not too washed out. In this case, I just recycled settings from my last Eta Carina image because I knew they worked.

When I do this again I will try for longer exposures, and maybe drop the UV-IR cut filter into the mix. According to the exposure time spreadsheet I've seen over on another (American) forum, I can bump it out to 20 second exposures.

The voices in my head are saying I should do two mosaics - one with the L-Enhance to boost the Ha and a second with the UV-IR cut and combine them. I'm trying to ignore what they're saying...

Oddly enough, I tend not to use the saturation controls when processing. This one was the exception to the rule. I've seen too many over-saturated images for my liking and don't like the way they look.

Cheers,
V

[Bodon (Steve)]

Steveo, looks brillant! Glad you have the weather. Im sure that you next will be a ripper.

[oska (John)]

I don't think you've over saturated, it's a very rich target. Working out what you like and then getting it is the struggle.
Listen to the voices, they're your wisdom, at least that's what mine said.