Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

15 Panel Mosaic, Aug 31st, 74% illumination


Xiga

Recommended Posts

So after waiting for what seemed like an eternity, I finally got around to trying my hand at a lunar mosaic. A few months back, I picked up a  (new to me) IDS camera cheaply on the 2nd hand market, and last Thursday (Aug 31st) I finally got my chance to put it through it's paces.

I used the C8 at it’s native focal length of 2032mm. The camera was an IDS-UI-6250SE-M/C GigE Mono camera, which uses a Sony ICX274 sensor (1600 x 1200, 4.4um, Global Shutter). So the resolution was just 0.45” (next time I might try using a x1.5 Barlow). I also used a basic cheap Wratten 23A red filter to help with the seeing somewhat (although with the moon being at just 10-15 degrees altitude I was also having to look through a ton of atmospheric murk!).

I picked the camera up for just £100, purely for Lunar work, and for the price I’d say I’m more than happy with it. It’s main drawback is the low fps (it can only manage 11 fps in 8 bit mode, and half that in 12 bit mode, crazy low by today’s standards!), but at least it’s a step up from my DSLR, and in any case I can just take longer captures to compensate, and having a Global Shutter should help as well. At some point in the future I'd like to get an ASI 178MM, but for now though I’m happy to just get my feet wet with this little guy.

So the main objective of the night was just to see if I could get all the software working ok, manage to set the exposure right and capture all the panels without missing any sections. I’m not sure how I did as regards the exposure (think the histogram was in the 50-60% range in FC, but I still ended up with overexposed areas), but at least the software side of things all went ok.

So this is a mosaic of 15 panels, captured with MoonPanoramaMaker (via Firecapture). The last 2 panels were captured as the moon was disappearing behind the neighbour’s roof, so I only had about 500 and 200 frames to work with on those. Just glad I managed to get something at all really, as it would have been very annoying to end up with a missing piece!
So due to the time constraints, I had to limit each tile to just 2 mins worth, so I chose the 8 bit mode in order to keep the number of frames up. In future I will use the 12 bit mode and just capture for longer. So each tile was just approx. 1,200 frames, from which I stacked the best 150 in Autostakkert3! I ran Maxim’s Deconvolution routine on each tile, then stitched them together using Microsot I.C.E. Then applied Wavelets in Registax6, followed by some final tweaks in Photoshop.

Due to the low number of frames it’s no surprise that it’s a bit noisy in places (I also probably pushed the sharpening a bit too far tbh). Also MS ICE couldn’t hide a few of the seams so I had to use some elbow grease to try and fix them (they still don’t look great).

Overall though I’m happy with how the test run went. Next time I should be a lot faster at setting up, polar aligning, focusing, using the software etc, so hopefully I can get longer captures to end up with a proper big mosaic that I can print off at a decent size.

Clear skies all!

36829789336_dbf6db7c52_h.jpg

 

 

Link to comment
Share on other sites

What happened toward the top of the image where the moon gets blurry?  Was that due to heat shimmers coming off the neighbor's roof?  Could you post one of your DSLR snaps for comparison to see the improvement?  The lower left edge near the terminator looks phenomenal.  It would be great if the whole image was that sharp (or was that just post-processing sharpening?).

Link to comment
Share on other sites

11 hours ago, Louis D said:

What happened toward the top of the image where the moon gets blurry?  Was that due to heat shimmers coming off the neighbor's roof?  Could you post one of your DSLR snaps for comparison to see the improvement?  The lower left edge near the terminator looks phenomenal.  It would be great if the whole image was that sharp (or was that just post-processing sharpening?).

Louis, thank you for bringing this to my attention! My initial reaction was that the top section would obviously have less contrast than the bottom, being more illuminated, but on closer inspection you are absolutely right, the processing was not consistent throughout (i don't have any DSLR photos as a comparison, i just meant that being a webcam it is just more suitable for the task in hand). 

So i went back to the drawing board and had another go. This time, i used Deconvolution on each individual panel (some several times over), rather than doing it on the whole mosaic. The Deconvolution worked so well (and kept the noise down amazingly well) that i didn't bother using Registax6 at all. In fact, from now on i think i'll stick solely to Deconvolution and avoid Regi alltogether. Oh, i also cropped each pane this time before putting them into MS ICE, and this really helped it to stitch them together so much better than before (i couldn't find any seams this time). 

New version attached below:

36846609836_22ede3dabf_h.jpg

Link to comment
Share on other sites

7 minutes ago, Xiga said:

So i went back to the drawing board and had another go. This time, i used Deconvolution on each individual panel (some several times over), rather than doing it on the whole mosaic. The Deconvolution worked so well (and kept the noise down amazingly well) that i didn't bother using Registax6 at all. In fact, from now on i think i'll stick solely to Deconvolution and avoid Regi alltogether. Oh, i also cropped each pane this time before putting them into MS ICE, and this really helped it to stitch them together so much better than before (i couldn't find any seams this time). 

Wow, big improvement!  Amazing what a difference post-processing techniques make today.  Keep up the good work.  I look forward to seeing more your work in the future.

Link to comment
Share on other sites

4 hours ago, Louis D said:

Wow, big improvement!  Amazing what a difference post-processing techniques make today.  Keep up the good work.  I look forward to seeing more your work in the future.

Thanks Louis!

Yes I agree, post processing is half the battle in this game :-)

Link to comment
Share on other sites

Ok made a few final tweaks, this time i really do think i'm done :tongue:

I fixed up the top portion a bit more, brought out a bit more detail towards the left of centre, and reduced the overall brightness as i had clipped quite a few areas without realising at the time. 

C&C most welcome. This was my first (hopefully of many) Lunar Mosaic attempts, so i'm looking for ways to improve for the next one. 

Quick question, i used a 23A red filter when taking the videos, but when i view them (in SER Viewer) and also when i processed them in AS3!, they show up as grayscale. Anyone have any idea as to why?

 36846609836_0a9d5f92ca_h.jpg

Link to comment
Share on other sites

4 minutes ago, Xiga said:

i used a 23A red filter when taking the videos, but when i view them (in SER Viewer) and also when i processed them in AS3!, they show up as grayscale.

I don't know anything about the camera, but could it be auto white balancing away the red tint?  Could it be capturing in monochromatic mode?

I like the fact that the ejecta from the crater at 11 o'clock is now fairly detailed.  I would recommend you pursue HDR (High Dynamic Range) imaging by decreasing your exposure for a certain percentage of frames and combining those that capture better detail in bright areas (such as the ejecta from the crater at 1:30) with the regular exposures.  I think there is software out there to help automate the merging process.

Overall, a magnificent image already.

Link to comment
Share on other sites

Hi Louis

 

I went back and looked at the .txt files from the Firecapture recordings, as they tell you what the avg Histogram level was. They ranged from about 50% up to the highest which was 93%. Interestingly though, the highest wasn’t the panel with the blown-out areas at the 1:30 position (I think it only had a level of about 75% so I’m surprised it ended up being so over-exposed). No it was the panel with the ejecta at the 11 o’clock position that had the 93% Histogram, yet when processed properly it actually doesn’t get blown-out (like you say I’ve managed this much better now, but I’ve still clipped it slightly, deliberately, just to bring the overall brightness level up). But no amount of processing can help with the 1:30 tile, it was way over-exposed to begin with somehow. So I’m a bit confused!

 

Your point about HDR imaging is interesting, I hadn’t actually considered that. My thought process (based on what I had researched) was basically, find the brightest area of the moon and set the exposure/gain such that it gives an avg histogram level of about 60-70% (which should still leave enough head-room during post-processing to avoid clipping), then don’t change it throughout the whole recording process, otherwise you could end up with uneven brightness levels across panels. But maybe modern processing techniques can deal with this ok, I’m just not sure tbh. I would love to find a really detailed, modern, guide to doing high resolution lunar mosaics, from acquisition right through to PP, but if there’s one out there then I haven’t found it! I see there’s a book called ‘Shoot the Moon’ on Amazon, I might take a punt on it and see if it’s any good.

 

As for the camera itself, it’s a mono cam. It always captures in mono mode so there’s no mucking about with debayering or white balance settings. It’s my first mono camera (and first of any kind that’s not a DSLR) so I’m new to all of this, but I would have thought that having the Red filter on there would still have resulted in Red captures. After all, that’s how the advanced imagers all do their work, with mono cams and filters, so what gives? I think I’ll drop the previous owner a quick email and see if he can think of anything.  

Link to comment
Share on other sites

12 hours ago, Xiga said:

It’s my first mono camera (and first of any kind that’s not a DSLR) so I’m new to all of this, but I would have thought that having the Red filter on there would still have resulted in Red captures

The wavelengths captured are all in red range due to the filtering, but the image is still B&W, just like in the old B&W film days.  If you shot images with blue and green filters as well, they would also be B&W.  However, each image would be slightly different.  You could combine the three images into a true color image by using the red image for the red channel, the blue for the blue channel and the green for the green channel.  It should remain mostly B&W, but if you stretch the curves, there is some color to the moon.

Link to comment
Share on other sites

2 hours ago, Louis D said:

The wavelengths captured are all in red range due to the filtering, but the image is still B&W, just like in the old B&W film days.  If you shot images with blue and green filters as well, they would also be B&W.  However, each image would be slightly different.  You could combine the three images into a true color image by using the red image for the red channel, the blue for the blue channel and the green for the green channel.  It should remain mostly B&W, but if you stretch the curves, there is some color to the moon.

Ahh, i see, i feel stupid now that i hear it said out loud, lol. This makes total sense now. 

Thanks for the explanation Louis! :thumbright:

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.