48MP 50% Lunar Mosaic (53 Panels) 02/03/2020

[matt_baker]

Imaging telescope: Celestron C14

Imaging camera: QHYCCD 5L ll M

Mount: Celestron CGE

Software: Autostakkert! 3,  PixInsight 1.8

53 Panels taken at 1280x960 30FPS. Stacked in AS!3, Deconvoluted and sharpened in PI, stitched together in ICE, then got rid of artefacts in PS

 

 

Edited March 4, 2020 by matt_baker

[barkingsteve]

53 Panels ?  are you a luna-tic lol   well done   

[matt_baker]

Haha yes, I was afraid that if I didn't overlap enough, then there was going to be a tiny gap somewhere so I ended up doing a lot. Could have been 10 less I reckon without the overlap

[Sunshine]

41 minutes ago, barkingsteve said:

53 Panels ?  are you a luna-tic lol   well done   

"Luna-tic" Very nifty, love it lol.

[paulastro]

So that's where I'm going wrong Mark - I need a C14 and a fancy camera 🤣.  Only joking Mark!  What fabulous imaging.  Thanks for posting. 🌝

[paulastro]

Ps sorry MATT!  I would have changed your name but my computer wouldn't let me edit it!

[matt_baker]

No worries haha and it's ironic because my dad was with me and his name is Mark

[matt_baker]

Done a few minor tweaks to make those craters just a tad sharper and increase the lightness

Please tell me what you think

[CraigT82]

That is an absolute stonker of an image!  53 panes must have taken you ages... how many frames did you capture/stack for each pane?

I prefer the first one as the second one looks a bit overdone on the sharpening when you zoom in (but looks great at overall)

[michael.h.f.wilkinson]

Great result. I do feel the sharpening is a bit too much to my personal taste, but only when I go pixel-peeping at full resolution. I prefer to use deconvolution and unsharp masking with ImPPG (automatically running on all panes).

[matt_baker]

Thanks for both of your replies!

I do agree that the second one is a little over sharpened when I pixel peep, especially around the terminator but I was thinking of using the second one as a photographic print instead

As for frames I captured 1000 frames per pane then stacked the best 75 in autostakkert

[Laurin Dave]

Very nice,  both are excellent and you absolutely do need to click through to see the high res version..  My preference is for version 1.  How long did the capture take and how do you execute it?

Dave

[matt_baker]

2 minutes ago, Laurin Dave said:

Very nice,  both are excellent and you absolutely do need to click through to see the high res version..  My preference is for version 1.  How long did the capture take and how do you execute it?

Dave

Thanks!

Overall it took me about an hour and a half to capture all of it. I used a rough 30% overlap with each pane to account for any error in tracking and for ICE. I achieved this using imerge and overlapping each frame to eventually create an entire mosaic. 

The school that I attend also has an in house observatory that houses the C14 and CGE and, although it's getting old, it does the job well and I can create images like this. I'm able to use the observatory whenever I want as I have a key and live a 5 minute walk away, so when it's clear, I can just hop over with my dad. It is a shame however, since I'm the only one who actively uses it because no one else either doesn't know how to operate it properly or they live too far away and the weather in England is always unpredictable

[matt_baker]

Going to re stack all the data with 1.5x drizzle and see how it goes. Gonna take me a while to do all 53 again lol

[Laurin Dave]

43 minutes ago, matt_baker said:

Going to re stack all the data with 1.5x drizzle and see how it goes. Gonna take me a while to do all 53 again lol

You might also experiment  with stacking a higher percentage of frames as this will bring the SNR up (so Damian Peach says  ) and allow mote aggressive sharpening

[matt_baker]

Maybe 125 this time?

[matt_baker]

Drizzling with 75 frames stacked per pane has made a huge difference, especially in terms of noise

Here's a comparison between the two (Left = Before Right = After)

Will try with 125 frames per pane next.

[CraigT82]

The image on the right is soooooo much better!! The drizzle will have helped as you are undersampled but I didnt think it would help that much! Did you sharpen the two images identically? 

[matt_baker]

7 minutes ago, CraigT82 said:

The image on the right is soooooo much better!! The drizzle will have helped as you are undersampled but I didnt think it would help that much! Did you sharpen the two images identically? 

Exactly what I was thinking! Yes, they're both the same (I think) or at least very very similarly edited. The one on the left may have had a tad more sharpening but nothing major.

Yes, it was undersampled but I never thought drizzling would make that much of a difference so I didn't bother with the extra time. Overall, a massive result!

[CraigT82]

Can't wait to see the finished article!

[matt_baker]

This is now the exact same processing. Two iterations of Decon in PI and HistogramTransformation on same settings

[matt_baker]

This is a better view actually

[geoflewis]

Hi Matt, I'm only now just seeing this thread and so glad that I came across it. It's a superb mosaic and I agree with others that the 1st version is best, the 2nd one is just over sharpened for my taste, but that might work very well for a poster print rather than pixel peeping. The drizzle looks to be working well based on the comparison frames, so it would definitely be worth working up the full mosaic, which I look forward to seeing.

[gorann]

A stunning image and like everyone else I am looking forward to the new drizzled mosaic. After that you just have to shoot the other half of the moon😉

[matt_baker]

22 minutes ago, gorann said:

A stunning image and like everyone else I am looking forward to the new drizzled mosaic. After that you just have to shoot the other half of the moon😉

If only UK weather will permit that 🤣