Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Saturn on 2019 August 17 from SW Oregon

  • Please log in to reply
8 replies to this topic

#1 Rustler46

Rustler46

    Viking 1

  • *****
  • topic starter
  • Posts: 938
  • Joined: 10 Feb 2018
  • Loc: Coos Bay, Oregon

Posted 19 August 2019 - 01:52 AM

Seeing was pretty poor last night. But at least I could see the Cassini Division on Saturn to focus on. With a fairly tight region of interest (ROI) I was able to capture 20,000 frames in an AVI video of less than 4 GB. To keep the planet's disc near the center of the ROI, I used the mount's control paddle to make adjustments for tracking or seeing induced movement. Using AutoStakkert!-2 a thousand of the best frames were stacked and fed into Registax-6. After some mild processing in Photoshop Elements-2 and Lightroom-5, here's the result:

 

Screen Shot 2019-08-18 at 10.58.46 PM.png

Celestron-11, f/17 (4750 mm f.l.), ZWO ASI290MC, ZWO ADC, Celestron 2X Barlow lens set on camera nosepiece, Losmandy G-11

 

This screen shot is a little brighter and larger:

Screen Shot 2019-08-19 at 12.59.50 AM.png

 

There was an odd set of 3 vertical lines on the inside edge of the B ring extending into where the C ring would be. This was only apparent on the right side of Saturn's disc. So I used the Photoshop Elements-2 smudge-tool to make this defect less visible.

 

Saturn was really jumping around and warping. But there were enough sharp frames to work with. I didn't bother adjusting the ADC except for giving it the maximum correction. In the past for such low elevation objects, the maximum correction did not completely remove dispersion. But it was a great improvement. With Saturn's disc at 17.9 arc-seconds and just over 23° elevation, conditions were challenging. So I was pleased to have gotten what you see. 

 

I'm beginning to get the cobwebs dusted off as I get back into planetary imaging. With Mars coming around next year, conditions will be less challenging, at least as far as elevation. So any experience gotten this year will be useful down the road.

 

Best Regards,

Russ


Edited by Rustler46, 19 August 2019 - 03:03 AM.

  • Jeffrey C., frank5817, Kenny V. and 5 others like this

#2 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3896
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 19 August 2019 - 05:05 AM

The vertical lines sound like a stacking artifact. Try playing around with the size and number of alignment points (boxes) in Autostakkert. I find the multi-scale option often works well too.

 

Great images though, very nice. I wonder if you could be a little more robust with the wavelets in registax, but lovely regardless.


  • RedLionNJ likes this

#3 RedLionNJ

RedLionNJ

    Soyuz

  • *****
  • Posts: 3661
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 19 August 2019 - 01:09 PM

The vertical lines sound like a stacking artifact. Try playing around with the size and number of alignment points (boxes) in Autostakkert. I find the multi-scale option often works well too.

 

Great images though, very nice. I wonder if you could be a little more robust with the wavelets in registax, but lovely regardless.

Both my thoughts, too, Nick.

 

I would try stacking maybe 5000 frames and use auto-placement for fairly small APs (including close to the edge).

 

See what comes of it....



#4 Rustler46

Rustler46

    Viking 1

  • *****
  • topic starter
  • Posts: 938
  • Joined: 10 Feb 2018
  • Loc: Coos Bay, Oregon

Posted 20 August 2019 - 10:04 PM

Thanks for the tips happylimpet and RedLionNJ. I've made another go at processing the videos using your suggestions. Alignment boxes in AutoStakkert!-2 were reduced in size to 24, automatically placed. This resulted in over 200 alignment points, some encompassing the edge of the disc and rings. In Registax-6 I tried all manner of processing (linear or Dydadic and default or Gaussian). Along with the various wavelet sliders there are a tremendous number of permutations. I found the 2nd and 3rd wavelet sliders to be most useful. Each slider has 3 potential adjustments (including noise reduction, etc.) - again many possibilities to explore.

 

I tried stacking 5000 of 20,000, 2000/20,000, 750/20,000 2000/10,000 for the two videos available. Nothing seemed to make a major improvement over first attempts. If anything there was a slight increase in detail shown on the disc. But there was no sign of the vertical lines present before. So they were some sort of stacking artifact that went away by using your suggestions.

 

Screen Shot 2019-08-20 at 6.56.44 PM.png

 

Stacked 2000 of 20,000 frames.

 

But still a soft image commensurate with the poor seeing when the video was captured. Tonight I'll have another brief opportunity between a cloudy day, clearing near sunset and a minor summer cold front. I'm hoping if it stays clear until Saturn is transiting that seeing might be tolerable. At least the day/night temperature differential is reduced with  mostly cloudy conditions during daylight hours.

 

Edit:

Last night provided a brief break in the clouds for some imaging of Saturn. Over the course of a couple of hours the seeing went from terrible to bad, then poor, then brief intervals of fair to good. So I was able to focus on the Cassini Division and gather 27 videos of 15,000 frames each. This amounted to 87 GB on the SSD. If anything of value results from these, I'll share the results.

 

Thanks again for the encouragement and suggestions. These are much appreciated.

 

Best Regards,

Russ


Edited by Rustler46, 21 August 2019 - 09:26 AM.

  • Kenny V. and happylimpet like this

#5 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3896
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 21 August 2019 - 09:21 AM

200 is probably overkill! I would have thought something like 30, and then selecting the multiscale option, should do it.

 

Having too many APs in areas with insufficient detail for them to 'lock on' harms the result.

 

My recommendation is to use the percentage boxes when stacking and select 5%, 12%, 30% and 75%, then you can easily compare. Also select the 'conv' output tickbox option, then flick through the 'conv' output files (you can search for these in your output folder by entering conv in the searchbar at top right) to quickly find the best to process properly in registax.

 

I usually find the 12% stack is the best for colour images.


  • Rustler46 likes this

#6 Rustler46

Rustler46

    Viking 1

  • *****
  • topic starter
  • Posts: 938
  • Joined: 10 Feb 2018
  • Loc: Coos Bay, Oregon

Posted 21 August 2019 - 01:35 PM

200 is probably overkill! I would have thought something like 30, and then selecting the multiscale option, should do it.

 

Having too many APs in areas with insufficient detail for them to 'lock on' harms the result.

 

My recommendation is to use the percentage boxes when stacking and select 5%, 12%, 30% and 75%, then you can easily compare. Also select the 'conv' output tickbox option, then flick through the 'conv' output files (you can search for these in your output folder by entering conv in the searchbar at top right) to quickly find the best to process properly in registax.

 

I usually find the 12% stack is the best for colour images.

Thanks for the comments, happylimpet! Advice from experienced planetary imagers is a great help. I'll try using your suggestions. One thing that is a factor in available options is that I'm using AutoStakkert!-2, rather than the latest AutoStakkert!-3. Here are some screen shots of the interface:

 

Screen Shot 2019-08-21 at 11.11.39 AM.png

 

Screen Shot 2019-08-21 at 11.13.03 AM.jpg

 

I don't see where it's possible to use "multi scale option" and "conv output". Maybe in AS-3 it's more obvious. I did notice one of my output files had conv in the file name. So the conv output is hidden there somewhere. But I have no idea what enabled it. 

 

In any case for the time being I'll be using AutoStakkert!-2. But eventually I'll need to enable AS-3. Soon the Mac OS will no longer support 32-bit programs.

 

Best Regards,

Russ



#7 happylimpet

happylimpet

    Soyuz

  • *****
  • Posts: 3896
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 21 August 2019 - 02:30 PM

Thanks for the comments, happylimpet! Advice from experienced planetary imagers is a great help. I'll try using your suggestions. One thing that is a factor in available options is that I'm using AutoStakkert!-2, rather than the latest AutoStakkert!-3. Here are some screen shots of the interface:

 

I don't see where it's possible to use "multi scale option" and "conv output". Maybe in AS-3 it's more obvious. I did notice one of my output files had conv in the file name. So the conv output is hidden there somewhere. But I have no idea what enabled it. 

 

In any case for the time being I'll be using AutoStakkert!-2. But eventually I'll need to enable AS-3. Soon the Mac OS will no longer support 32-bit programs.

 

Best Regards,

Russ

I'd get AS!3, but I dont think it makes a huge difference. Presumably the 'multiscale' option is an AS!3 thing! Its on the bottom left of the lower window, a tick box close to "Place AP grid".

 

If you tick 'sharpened' you get the _conv files also. I strongly recommend it. SOmetimes theyre as good as your final Registax'ed version! Or almost.

 

Also, tick the 'normalize' option. It stops you getting dark final images, and if you used different capture settings it makes all the images look the same brightness, which is handy.


  • Rustler46 likes this

#8 Tulloch

Tulloch

    Messenger

  • -----
  • Posts: 486
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 21 August 2019 - 06:15 PM

I'd get AS!3, but I dont think it makes a huge difference. Presumably the 'multiscale' option is an AS!3 thing! Its on the bottom left of the lower window, a tick box close to "Place AP grid".

 

If you tick 'sharpened' you get the _conv files also. I strongly recommend it. SOmetimes theyre as good as your final Registax'ed version! Or almost.

 

Also, tick the 'normalize' option. It stops you getting dark final images, and if you used different capture settings it makes all the images look the same brightness, which is handy.

Yep, the multiscale option is an AS!3 feature. +1 on the normalize option too, this should always be ticked. I found that once I worked out some good Registax settings, I didn't need to create a "sharpened" version from AS!. However, this might be a good option while still learning the tools.

 

One thing I always use is the Drizzle 3x option. Sure, it takes 3x longer to process the files, but you get an image 3x larger which I find helps with sharpening and doing any final corrections. Once you've made these enhancements to the larger file, you can then reduce down by 33% to get an image at capture scale (although I usually reduce by 50% to give a slightly bigger image). Your Registax settings need to change (make them larger) to cope with the bigger images, but I found it's certainly worth it. 

 

Registax also has a feature to auto-balance RGB, which usually produces a pretty good colour scheme which you can then work on in PS, LR etc.

 

I may have said this before, but Steve's videos on how to use these programs are very good, well worth a look, especially the stacking in AS!2 and processing in Registax videos.

 

http://planetaryimag...tutorials.com/ 

 

Hope this helps,

 

Andrew


Edited by Tulloch, 21 August 2019 - 06:16 PM.

  • Rustler46 likes this

#9 Rustler46

Rustler46

    Viking 1

  • *****
  • topic starter
  • Posts: 938
  • Joined: 10 Feb 2018
  • Loc: Coos Bay, Oregon

Posted 22 August 2019 - 12:02 AM

Yep, the multiscale option is an AS!3 feature. +1 on the normalize option too, this should always be ticked. I found that once I worked out some good Registax settings, I didn't need to create a "sharpened" version from AS!. However, this might be a good option while still learning the tools.

 

One thing I always use is the Drizzle 3x option. Sure, it takes 3x longer to process the files, but you get an image 3x larger which I find helps with sharpening and doing any final corrections. Once you've made these enhancements to the larger file, you can then reduce down by 33% to get an image at capture scale (although I usually reduce by 50% to give a slightly bigger image). Your Registax settings need to change (make them larger) to cope with the bigger images, but I found it's certainly worth it. 

 

Registax also has a feature to auto-balance RGB, which usually produces a pretty good colour scheme which you can then work on in PS, LR etc.

 

I may have said this before, but Steve's videos on how to use these programs are very good, well worth a look, especially the stacking in AS!2 and processing in Registax videos.

 

http://planetaryimag...tutorials.com/ 

 

Hope this helps,

 

Andrew

Thanks very much Andrew for the further tips. I'm beginning to work through the 27 videos captured last night. At first I was a bit disappointed at the results. Then I recalled that the seeing was improving. So working with some videos later in the evening looks much more promising. I'll post some results in the course of time.

 

Best Regards,

Russ




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics