#1  
Old 20-09-2020, 09:50 AM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Winjupos

I'm starting to use this trying to derotate low SNR Saturn SER files. My understanding is that you create an image from a video stack with AS!3 as per usual, measure a feature on the planet then derotate the video streams (let's say 3min SER) and create a new SER output from winjupos.

My question is do I create the still picture from the first (3min SER) or from the joint 3 files (9min SER)?

AS!3 will automatically sort the frames in quality order prior to stack so I will end up in the first 25% with a mix of the 3 videos that have frames already rotated if I use the whole 9min in one file.

My thinking is that for the reference image in Winjupos you'd want something that is not already blurred right?
Reply With Quote
  #2  
Old 21-09-2020, 01:11 PM
Tulloch's Avatar
Tulloch (Andrew)
Registered User

Tulloch is offline
 
Join Date: Sep 2019
Location: Melbourne
Posts: 488
Hi Marc, you seem to be a bit muddled here.

There are two ways to derotate in WinJupos, either by image or using the video. Image derotation is the easiest (if you are on an EQ mount), video derotation is the best way if you are on an alt/az mount.

For image derotation, you stack and sharpen the image from the video in AS!3 and Registax, perform the "Image Measurement" steps on each image, then use the "Image derotation" steps in the "Tools" menu. This just derotates the images, putting the features on top of each other and so improving the final image.

For video derotation, you need to select the best frame from your raw video, do the "Image measurement" step as before, then use this as the reference frame for the video in the "Derotation of video streams" in the Tools menu. WJ will then derotate every frame in the video (both planetary rotation and field rotation if you ask it to do so), creating another video which you then put into AS!3 for stacking. This is far more time consuming than image derotation.

Here are some tutorials, I'm sure there are others about...
https://www.skyatnightmagazine.com/a...netary-images/
https://www.youtube.com/watch?v=baIgfj5eqc8

Andrew
Reply With Quote
  #3  
Old 21-09-2020, 02:13 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Thanks Andrew, will do some more reading.
Got told best to stack small videos, get best subs and derotate a list of still pictures.
Reply With Quote
  #4  
Old 21-09-2020, 06:41 PM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Hi Marc,

Have been using WinJupos for several years and I never de-rotate videos.

Simply stack your frames in AS3 but only take videos long enough which do not cause rotational issues. e.g. for me it's about 100 seconds for Jupiter and 120 seconds for Mars.

Then stack as per normal in AS3, do the image measurement in WJ - if you like you can de-rotate several images in WJ for individual R,G,B channels or several colour channels - then if mono, process your individual de-roated R, G, B, files, do image measurements again of these in WJ, and finally combine in WJ which will de-rotate and combine them into an RGB or LRGB.

WJ is a very powerful tool which has allowed amateur imagers to start processing at a professional level - recently Chris Go was live on YouTube and the Hubble guys are going to get him to process some of their planetary images with WJ as he can get more details out of the raw hubble data on Saturn than they can!

Hope it all makes sense!

John K.
Reply With Quote
  #5  
Old 21-09-2020, 07:05 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by John K View Post
Simply stack your frames in AS3 but only take videos long enough which do not cause rotational issues. e.g. for me it's about 100 seconds for Jupiter and 120 seconds for Mars.
That's pretty short in time. How many frames do you get in 100s on Jupiter and 120s on Mars?

Quote:
Originally Posted by John K View Post
Then stack as per normal in AS3, do the image measurement in WJ - if you like you can de-rotate several images in WJ for individual R,G,B channels or several colour channels - then if mono, process your individual de-roated R, G, B, files, do image measurements again of these in WJ, and finally combine in WJ which will de-rotate and combine them into an RGB or LRGB.
Thanks for the explanation. Taking notes. I will try that next.

I attached a file I joined (3x3min in PIPP) to make one SER derotated stream of Mars in WinJUPOS. There is an obvious line due to the process although the overall details are better than the one 3min stream stacked.
Attached Thumbnails
Click for full-size image (Mars_003137_L_120920_AS_P88_lapl5_ap46.jpg)
61.4 KB57 views
Reply With Quote
  #6  
Old 21-09-2020, 07:21 PM
Nikolas's Avatar
Nikolas (Nik)
Dazed and confused

Nikolas is offline
 
Join Date: May 2012
Location: Melbourne
Posts: 3,253
Best and easiest tutorial for winjupos out there by our very own Dylan
https://www.youtube.com/watch?v=i5BdwYpWW-s
Reply With Quote
  #7  
Old 21-09-2020, 07:42 PM
Tulloch's Avatar
Tulloch (Andrew)
Registered User

Tulloch is offline
 
Join Date: Sep 2019
Location: Melbourne
Posts: 488
Quote:
Originally Posted by multiweb View Post
I attached a file I joined (3x3min in PIPP) to make one SER derotated stream of Mars in WinJUPOS. There is an obvious line due to the process although the overall details are better than the one 3min stream stacked.
If your videos aren't continuous (ie the next recording started directly after the previous one finished), then you shouldn't use the video derotation method, it's far better to create separate images from each video then use the Image derotate method.

It's only recently that I've started using WJ to derotate videos, since I use an alt/az mount the image derotate method is far more problematic due to slight misalignment of the images caused by the field rotation.
Reply With Quote
  #8  
Old 22-09-2020, 09:22 AM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by Tulloch View Post
If your videos aren't continuous (ie the next recording started directly after the previous one finished), then you shouldn't use the video derotation method, it's far better to create separate images from each video then use the Image derotate method.

It's only recently that I've started using WJ to derotate videos, since I use an alt/az mount the image derotate method is far more problematic due to slight misalignment of the images caused by the field rotation.
Yeah they were continuous with a 3s gap between the sequences.
Makes sense on an Alt/Az. Reading up on WinJUPOS now.
Reply With Quote
  #9  
Old 22-09-2020, 01:19 PM
rustigsmed's Avatar
rustigsmed (Russell)
Registered User

rustigsmed is offline
 
Join Date: Mar 2012
Location: Mornington Peninsula, Australia
Posts: 3,950
WJ - is an absolute life-saver for alt-az / goto dobs.
Reply With Quote
  #10  
Old 22-09-2020, 03:05 PM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Quote:
Originally Posted by multiweb View Post
That's pretty short in time. How many frames do you get in 100s on Jupiter and 120s on Mars?
.
I’m averaging 8,000 to 10,000 based on 80 FPS.

Tend to stack about 30%.

John K.
Reply With Quote
  #11  
Old 22-09-2020, 03:38 PM
Tulloch's Avatar
Tulloch (Andrew)
Registered User

Tulloch is offline
 
Join Date: Sep 2019
Location: Melbourne
Posts: 488
Quote:
Originally Posted by rustigsmed View Post
WJ - is an absolute life-saver for alt-az / goto dobs.
Well, I've only just started using it, when I tried performing the image derotation the images always came out worse. I've noticed a slight difference in image quality using so far, but it's only slight ...
Reply With Quote
  #12  
Old 22-09-2020, 03:53 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by John K View Post
I’m averaging 8,000 to 10,000 based on 80 FPS.

Tend to stack about 30%.

John K.
Far out! You guys are smoking it!
30% out of 10000. I'm way greedier than you then. I stack 75% of my puny 4000 frames.

Although my fps has gone up lately. Haven't tried Mars again but last session I did 81fps in SER 8bit and 72fps in SER 16bit on Jupiter both in IR. I did 21fps in SER 8bit on Saturn and 17fps in SER 16bit. On the moon thin crescent I got up to 82fps in SER 8bit and 73fps in SER 16bit both in IR again.
Reply With Quote
  #13  
Old 23-09-2020, 11:13 AM
John K's Avatar
John K
Registered User

John K is offline
 
Join Date: Sep 2005
Location: Melbourne
Posts: 1,468
Quote:
Originally Posted by multiweb View Post
Far out! You guys are smoking it!
30% out of 10000. I'm way greedier than you then. I stack 75% of my puny 4000 frames.

Although my fps has gone up lately. Haven't tried Mars again but last session I did 81fps in SER 8bit and 72fps in SER 16bit on Jupiter both in IR. I did 21fps in SER 8bit on Saturn and 17fps in SER 16bit. On the moon thin crescent I got up to 82fps in SER 8bit and 73fps in SER 16bit both in IR again.
You can try some things - switch off the 16 Bit - no benefit in planetary; crank up the USB traffic to 80% or 90%; and also switch on High Speed.

The seeing in Melbourne is always variable so I find that 2,500 to 3,000 frames with the setting I have are more that sufficient to sharpen and process my images. But others might find this different. Experiment and find the optimal for your set up. But more frames = better results in my mind.
Reply With Quote
  #14  
Old 23-09-2020, 12:44 PM
Tulloch's Avatar
Tulloch (Andrew)
Registered User

Tulloch is offline
 
Join Date: Sep 2019
Location: Melbourne
Posts: 488
As John says, your frame rates are very low. In addition to what he mentioned, reduce your ROI as much as possible during capture and use high gain levels.

With my C9.25 and ASI224MC, my normal process is to image Jupiter for 3 minutes at 150 fps (so 27000 frames) at 280 gain from which I normally use stacks of 3000, 5000 or 7500 frames (seeing dependent). For Saturn I use 5 minutes at 100 fps (so 30000 frames total) at 370 gain, and stack 5000 to 10000 frames.

Last edited by Tulloch; 24-09-2020 at 06:57 AM.
Reply With Quote
  #15  
Old 23-09-2020, 01:20 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by John K View Post
You can try some things - switch off the 16 Bit - no benefit in planetary; crank up the USB traffic to 80% or 90%; and also switch on High Speed.
Yeah I have the USB traffic to the max and also high speed ticked on. I use 16bit although it's slightly slower on recommendations from Anthony (bird) who leaves it on all the time. It has a better dynamic range and there are many instances when you need the extra signal. I lose may be 10fps on it in worst case scenario. I guess it's all a compromise.

Quote:
Originally Posted by John K View Post
The seeing in Melbourne is always variable so I find that 2,500 to 3,000 frames with the setting I have are more that sufficient to sharpen and process my images. But others might find this different. Experiment and find the optimal for your set up. But more frames = better results in my mind.
I agree. The more the better. The stack difference between 4000 and 10000 definitely starts showing during processing and how much details you can get out and sharpen.

Quote:
Originally Posted by Tulloch View Post
As John says, your frame rates are very low. In addition to what he mentioned, reduce your ROI as much as possible during capture and use high gain levels.
Mate, any tighter and it will be tighter than a fish's ass. I have just enough black around the disc to keep the wobble within the FOV. I think the 120MM USB2.0 is an old camera. Maybe down the road if I get serious about planetary I'll get one of those flea cameras. Nearly bought one from the classified years ago. I should have.

Quote:
Originally Posted by Tulloch View Post
With my C9.25 and ASI224MC, my normal process is to image Jupiter for 3 minutes at 150 fps (so 27000 frames) at 280 gain from which I normally use stacks of 3000, 5000 or 7500 frames (seeing dependent). For Saturn I use 5 minutes at 100 fps (so 30000 frames total) at 37 gain, and stack 5000 to 10000 frames.
100fps on Saturn. That's my wet dream. Oh well, for now I'll do with what I have
Reply With Quote
  #16  
Old 23-09-2020, 06:23 PM
Tulloch's Avatar
Tulloch (Andrew)
Registered User

Tulloch is offline
 
Join Date: Sep 2019
Location: Melbourne
Posts: 488
Aaah, it's a USB-2 device, enough said. That will severely reduce the fps rate. The best bang-for-buck OSC planetary camera out there right now is either the just released ASI462MC ($499), older technology ASI224MC ($379), or at the bottom end is the ASI120MC-S ($239).

However, for planetary there is no benefit for running anything higher than 8 bit. There is a great explanation for that here, and increasing your framerate even a little bit will help more than capturing more noise.
https://www.cloudynights.com/topic/7...gs/?p=10313180
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 03:42 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement