M31, in glorious monochrome

[tomato]

This is the Lum part of my 12 panel mosaic heavily cropped, taken with the dual Esprit 150 rig and KAF 8300 CCDs. Each panel was 24 x 300 secs Lum, 8 x 300 secs each RGB.

This was supposed to be my 'proper' M31 Autumn project, first session was at the end of August, and it took 16 sessions to capture, finishing at the end of October. I started with the best of intentions, imaging when there was no moon, and capturing the blue channel when the galaxy was at it's highest. However, the dire UK weather put paid to that and I ended capturing a lot of panels with the moon both up and well illuminated, and I'm afraid I have paid the price. The channels have shocking gradients and have required very heavy handed techniques to remove them. I couldn't get DBE in PI to deliver an acceptable result, the best I could achieve was with APP's light pollution removal tool, after drawing no less than 6423 sample boxes, my mouse finger had RSI after that!

The colour channels are even worse, and my results to date have yielded either no colour to speak of, or it makes M31 look like one of those joke multicoloured wigs! Hopefully I can get something worth posting, but for now it is the Lum channel only. This is not great, there are some artefacts still in there, e.g. the dark 'spoke' at the rear of the galaxy? About the only benefit is the image can be zoomed in to a degree and still retain detail, but is it worth it?

Thanks for looking

[geeklee]

An admirable project in the UK without a doubt and I couldn't believe you'd still squeezed it in during that time period!  It was a real treat viewing at 1:1, especially around the core area and the dust lanes.  I guess only you can answer the best on whether it was worth it - does it meet your original goal(s)?

20 minutes ago, tomato said:

APP's light pollution removal tool, after drawing no less than 6423 sample boxes, my mouse finger had RSI after that!

Wow, that's a staggering amount of sample boxes! Apologies if you approached it like this already, but what about dealing with each panel independently - remove gradients, balance the backgrounds, then use APP to do the mosaic- assisted with LNC.  Did you test different LNC degrees and iterations in APP with the above?

Perhaps it can be a multi year project? <tumbleweed> Yeah, I know!😁

I'm always very impressed seeing longer term projects like this from challenging skies & weather - this was no different 👍

[tomato]

1 minute ago, geeklee said:

An admirable project in the UK without a doubt and I couldn't believe you'd still squeezed it in during that time period!  It was a real treat viewing at 1:1, especially around the core area and the dust lanes.  I guess only you can answer the best on whether it was worth it - does it meet your original goal(s)?

Wow, that's a staggering amount of sample boxes! Apologies if you approached it like this already, but what about dealing with each panel independently - remove gradients, balance the backgrounds, then use APP to do the mosaic- assisted with LNC.  Did you test different LNC degrees and iterations in APP with the above?

Perhaps it can be a multi year project? <tumbleweed> Yeah, I know!😁

I'm always very impressed seeing longer term projects like this from challenging skies & weather - this was no different 👍

Thanks, I'm experimenting with different LNC and MBB settings as I type, at least with the UK weather I have lots of time to do this. I haven't tried applying LP control to each panel then building the mosaic from these, it does make sense so I will give it a try.

Would I embark on another large scale mosaic any time soon? The answer has to be no, but the fact that I have recently  purchased a RASA 8 and QHY 268c has quite a bit to do with my decision.

[Laurin Dave]

 Nice one ... I’d suggest that part processing  each panel before combining as a mosaic might help..  in particular remove gradients from each panel first and get the backgrounds at the same level first . Also a light stretch may help..  Then when you do the mosaic set it up so that as well as producing a full mosaic APP also produces the separate panels ..  These will facilitate Photoshop repair of artefacts ..

HTH

Dave 

[gorann]

Admirable attempt Steve! I do not have the paitence (and weather) for mosaics larger than 2 - 4 panels, so this is really impressive. Maybe you can collect the colour with the RASA - will be done in a few hours I promise😉 and then combine it with your lum. Is it possible to reduce the brightness of the galactic cores a bit?

[maw lod qan]

Very nice image!

Every time I see it imaged I try to picture what it would have been like to look up at it 300 years ago when the only light pollution came from candles and campfires. 

[tomato]

18 minutes ago, gorann said:

Admirable attempt Steve! I do not have the paitence (and weather) for mosaics larger than 2 - 4 panels, so this is really impressive. Maybe you can collect the colour with the RASA - will be done in a few hours I promise😉 and then combine it with your lum. Is it possible to reduce the brightness of the galactic cores a bit?

Yes, this idea had occurred to me if I have to write off the RGB CCD panels. Perhaps I could get a decent M31 @ 1.94 arcsec per pixel in it's own right, and generate the colour channels for the mosaic all in one session.? The RASA is going back on the Mesu tomorrow...

[gorann]

The very best of luck with the RASA Steve! Does it mean it is going to be clear in your corner tomorrow? To remind you and maybe inspire you, this is what my RASA 8 and ASI2600 did during its first light, aimed on M31 during 2.6 hours back in August. The image is a bit low in the red so maybe I should collect some M31 Ha with the RASA if it would clear here too.....

It is 1.94 "/pixel as you say but it is still a quite ok resolution so I now had a closer look at the core (with HiPass filtering in PS) and I think the resolution is good enough for being a wide field image of M31.

[tomato]

Yes, I'll be very happy if I can get close to that. I set up the scope tonight, I have a Celestron auto focuser on it which works really well, I tried it out with NINA, all connected great but I could not get the camera to loop, need to read the manual!

Boy is the camera sensitive, with a 90% plus illuminated moon, 45 sec exposures are washed out within 50 deg of it, but it's on the wane.👍

[gorann]

1 minute ago, tomato said:

Yes, I'll be very happy if I can get close to that. I set up the scope tonight, I have a Celestron auto focuser on it which works really well, I tried it out with NINA, all connected great but I could not get the camera to loop, need to read the manual!

Boy is the camera sensitive, with a 90% plus illuminated moon, 45 sec exposures are washed out within 50 deg of it, but it's on the wane.👍

You are lucky! I cannot even see any moon tonight🥴 By the way, I have found focussing the RASA8 very simple - do not need any special device, and it holds it all night.

[messier 111]

fascinating take .

[ollypenrice]

I wouldn't expect any software currently available to be able to sort out a big mosaic by throwing the raw data into the pot and pressing 'Go.' I think that, as things stand, mosaics are partly automated, partly hand made. I had the great good fortune to be invited to play with Yves Van den Broek's 32 panel extravaganza around Cepheus.  https://www.astrobin.com/g82xf7/B/?nc=user

Let's look at the software: Yves tried combining the lot in APP. In some was the result was superb, in some ways not. APP did give a geometrically coherent picture. Because he'd used plate solving and a mosaic maker I felt that the data ought to be pretty 'true to the sky' and the output mosaic was almost oblong but not quite. Rightly or wrongly (I'm out of my depth) I used Ps to remove the slight distortion and made it oblong. This was a small adjustment but let me keep some nice bits in the final oblong crop.

Once I had this geometric template I had to make each panel fit it seamlessly. Ah! The fun bit. 😁 I used a combination of Pixinsight, Registar and Photoshop and got as far as I got. You can see the image. Starting again I could do better but, hey, 32 panels!!!

The day will come when the highly stretched astrophographic mosaic will be a one click affair but we are not there yet. Or, at least, I'm not!

Olly

Edited November 4, 2020 by ollypenrice

[Phillyo]

That mosaic is absolutely stunning Olly. From a OSC camera too! I was initially concerned I'd made a mistake with my ASI533 but seeing images like this, albeit out of my capabilities, gives one hope for the OSC cameras moving forward.

[ollypenrice]

31 minutes ago, Phillyo said:

That mosaic is absolutely stunning Olly. From a OSC camera too! I was initially concerned I'd made a mistake with my ASI533 but seeing images like this, albeit out of my capabilities, gives one hope for the OSC cameras moving forward.

I think CMOS has changed OSC.

Olly