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 20, Elevation 23°

  • Please log in to reply
21 replies to this topic

#1 Rustler46

Rustler46

    Viking 1

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

Posted 22 August 2019 - 05:25 PM

Here is a preliminary result from a successful imaging run the other night on Saturn, our ringed wonder. Over the course of some 2 hours there were moments of fair to good seeing. The resulting 27 videos of around 2-1/2 minutes contained 15,000 frames each. Depending on gain, the frame rate was around 95 fps, but sometimes as low as 75 fps. Here's the equipment used:

  • Celestron-11, f/17, 4750 mm focal length, Celestron 2X Barlow lens set on nosepiece of camera
  • ZWO ASI290MC camera
  • ZWO ADC, atmospheric dispersion corrector set at maximum correction
  • MacBook Pro computer (mid-2012), 16 GB memory, 500 GB SSD, USB 3.0

At the time of capture Saturn's disc had an angular extent of 17.8 arc-seconds. Sampling was 0.13 arc-seconds per pixel. The resulting 87-1/2 GB of video is taking a while to sort through and process. Early on the seeing was terrible. But some of the later videos are promising. Yet the low elevation of around 23° limited what could be captured. Also the relatively brief periods of good seeing limited the number frames of sufficient quality to include in the stack. Still I was pleased with some preliminary results.

 

Here is the work flow:

  • Image capture via FireCapture 2.6
  • Stacking with AutoStakkert!-2
  • Wavelet sharpening by Registax-6
  • Final adjustment by Adobe Lightroom-5

Following some very useful advice by forum members, each video had the best frames stacked at the 5%, 12%, 30% and 75% levels. Sharpening and RGB align were applied in AutoStakkert!-2. As yet drizzle hasn't been explored. But that will be applied once the better videos are identified.

 

Registax-6 is always an adventure, exploring the multitude of options available. Of the avenues most often used were "linear" and "Gaussian". Wavelets 2 and 3 were most useful, along with wavelet 1 for the best stacks. RGB align was again employed, along with appropriate noise reduction in each wavelet.

 

So here is a preliminary result:

 

Screen Shot 2019-08-22 at 2.40.25 PM.png

 

C-11, ASI 290MC, ADC, best 5% (750) of 15,000 frames, stacked with AutoStakkert!-2, sharpened with Registax-6

 

There was a tradeoff between fewer, sharper frames stacked producing a noisier image. I'll share some other results from the 12% and 30% stacks when these have been completed. The larger number of frames showed less detail, but a smoother image, particularly on Saturn's disc. One continuing problem was a stacking and sharpening artifact where the ring went behind the disc on the sunlit side. 

 

There will be more results as I have time to further investigate the 87 GB of videos.

 

Regards,

Russ


Edited by Rustler46, 22 August 2019 - 07:35 PM.

  • ozzmosis, Sunspot, frank5817 and 10 others like this

#2 Peter L.

Peter L.

    Ranger 4

  • -----
  • Posts: 348
  • Joined: 19 Apr 2008
  • Loc: PEI, Canada

Posted 22 August 2019 - 05:57 PM

Hey Russ,

 

Very nice. I can see a bit of the Hex.

 

Question for you - why the barlow right in front of the camera? Typical practice is for the barlow to be out in front, and the ADC between the barlow and camera. Are you trying to keep the F/L down around 15-17 for your 290MC camera? Others here can explain better than I why barlow - ADC - camera is preferred over ADC - barlow - camera.



#3 Rustler46

Rustler46

    Viking 1

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

Posted 22 August 2019 - 06:14 PM

Hey Russ,

 

Very nice. I can see a bit of the Hex.

 

Question for you - why the barlow right in front of the camera? Typical practice is for the barlow to be out in front, and the ADC between the barlow and camera. Are you trying to keep the F/L down around 15-17 for your 290MC camera? Others here can explain better than I why barlow - ADC - camera is preferred over ADC - barlow - camera.

Thanks for the encouraging comment, Peter. Oh yeah, I hadn't noticed the indication of polar Hex. Maybe it will be better displayed as I get some of the larger stacks processed.

 

As for the placement of the Barlow, with the tiny 2.9μ pixels of the ASI290MC, I didn't need a large focal length. Even at 4750 mm the sampling is quite high at 0.13 arc-seconds per pixel. But I'm sure open to suggestions as to proper use of the equipment. In my maritime environment, seeing isn't usually conducive to a long focal length. Next year's Mars conjunction will be a nice opportunity to further explore planetary imaging.

 

Best Regards,

Russ



#4 Peter L.

Peter L.

    Ranger 4

  • -----
  • Posts: 348
  • Joined: 19 Apr 2008
  • Loc: PEI, Canada

Posted 22 August 2019 - 06:37 PM

Thanks for the encouraging comment, Peter. Oh yeah, I hadn't noticed the indication of polar Hex. Maybe it will be better displayed as I get some of the larger stacks processed.

 

As for the placement of the Barlow, with the tiny 2.9μ pixels of the ASI290MC, I didn't need a large focal length. Even at 4750 mm the sampling is quite high at 0.13 arc-seconds per pixel. But I'm sure open to suggestions as to proper use of the equipment. In my maritime environment, seeing isn't usually conducive to a long focal length. Next year's Mars conjunction will be a nice opportunity to further explore planetary imaging.

 

Best Regards,

Russ

I have a similar setup as you, Russ, but with a C14. I agree that F/20 is just too much for my skies as well, so I use a 1.5x barlow. My setup is Filter Wheel --> Barlow --> ADC --> Camera. The barlow lens is cut from its casing and placed inside a T2 and slid back towards the ADC as far as it will go. This gives me about F/17 on my F/11 C14. I use a 290MC with this config as well. When I go to the 2x barlow...which is very, very, rare, I switch to the 224MC because I'm up around F/20 then.

 

Again, wish I could explain the advantages of having the barlow in front of the ADC, but I believe there are some. Hopefully someone who knows interjects.


Edited by Peter L., 22 August 2019 - 06:38 PM.


#5 Rustler46

Rustler46

    Viking 1

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

Posted 22 August 2019 - 07:32 PM

I have a similar setup as you, Russ, but with a C14. I agree that F/20 is just too much for my skies as well, so I use a 1.5x barlow. My setup is Filter Wheel --> Barlow --> ADC --> Camera. The barlow lens is cut from its casing and placed inside a T2 and slid back towards the ADC as far as it will go. This gives me about F/17 on my F/11 C14. I use a 290MC with this config as well. When I go to the 2x barlow...which is very, very, rare, I switch to the 224MC because I'm up around F/20 then.

 

Again, wish I could explain the advantages of having the barlow in front of the ADC, but I believe there are some. Hopefully someone who knows interjects.

Thanks for the update on your equipment and location. First of all, what a beautiful place where you reside. Light pollution must be a lot better on Prince Edward Island than most of us endure. I've always wanted a C-14, but only with a permanent observatory. It's just too heavy for me to lift repeatedly. You are located 3° further north than myself. So Saturn is that much lower in the sky. Just one of many challenges we face.

 

I'd like to hear what others say about Barlow placement in the imaging train. If better on the other side of the ADC, I'll likely go to a 1.5X version, since f/17 seems to be about right for sampling.

 

All the Best,

Russ



#6 Peter L.

Peter L.

    Ranger 4

  • -----
  • Posts: 348
  • Joined: 19 Apr 2008
  • Loc: PEI, Canada

Posted 23 August 2019 - 05:21 AM

Hi Russ,

 

I went back through some old posts to see if I could find the answer, but found lots of contradictory opinions instead. Looks like some are using the ADC - barlow - camera setup same as you and achieving good results, so perhaps there is no advantage to changing after all? Perhaps I had just convinced myself that there was? Sorry for misleading.


Edited by Peter L., 23 August 2019 - 05:21 AM.

  • Rustler46 likes this

#7 kevinbreen

kevinbreen

    Surveyor 1

  • -----
  • Posts: 1852
  • Joined: 01 Mar 2017
  • Loc: Wexford, Ireland

Posted 23 August 2019 - 06:21 AM

The further the Barlow is from the sensor the higher the focal length will be and hence also the focal ratio. If you can imagine the light rays spreading out further with the increased distance then you’ve got it. They’re spread out over a wider area on your sensor and the image appears bigger, but the light intensity is lower and darker, hence the need for longer exposure time and higher gain, which is often counter-productive.

My image train is like this

F/SCT - ir cut filter - ZWO ADC - 2X Barlow lens element - ASI224MC

and it gives me around f/17 ish.

Edited by kevinbreen, 23 August 2019 - 06:24 AM.


#8 RedLionNJ

RedLionNJ

    Soyuz

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

Posted 23 August 2019 - 07:48 AM

I believe the primary reason for maximizing the focal ratio PRIOR to light entering the ADC is that you can achieve a larger magnitude of dispersion correction, in proportion to the focal ratio.

 

Realistically, as long as you're able to adjust your ADC optimally, I believe that's all that matters.  Don't forget the mere use of the ADC (with the levers anywhere but null) is already introducing another sort of aberration, but it's still smaller than the actual dispersion you're correcting for.

 

My opinion has always been to optimize the f-ratio and let the ADC end up wherever it needs to be to get the desired f-ratio. Perhaps if i had perfect seeing, I could detect a difference between the two placements. I don't anticipate that ever happening.


  • Rustler46 likes this

#9 Rustler46

Rustler46

    Viking 1

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

Posted 23 August 2019 - 06:04 PM

There will be more results as I have time to further investigate the 87 GB of videos.

I disposed of around 40 GB of videos by quickly analyzing them in AutoStakkert!-2. Starting with the earlier ones (when seeing was garbage) if there wasn't a significant fraction of frames above the 75% quality level - these were sent to the trash. Still more to cull out.

 

Russ



#10 Rustler46

Rustler46

    Viking 1

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

Posted 23 August 2019 - 06:28 PM

I believe the primary reason for maximizing the focal ratio PRIOR to light entering the ADC is that you can achieve a larger magnitude of dispersion correction, in proportion to the focal ratio.

 

Realistically, as long as you're able to adjust your ADC optimally, I believe that's all that matters.  Don't forget the mere use of the ADC (with the levers anywhere but null) is already introducing another sort of aberration, but it's still smaller than the actual dispersion you're correcting for.

 

My opinion has always been to optimize the f-ratio and let the ADC end up wherever it needs to be to get the desired f-ratio. Perhaps if i had perfect seeing, I could detect a difference between the two placements. I don't anticipate that ever happening.

Thanks for the information on use of the ADC. At f/17 on the C-11 (Barlow just before camera) the ADC doesn't quite remove all dispersion at maximum correction. But the remaining error is inconsequential. Just doing an RGB align in processing removes enough remaining dispersion for my tastes. This simplifies the work flow. Just align the ADC with horizon, set correction to maximum and acquire the planet. Maybe as I become more proficient at planetary imaging, the issue of getting perfect dispersion correction will be addressed. Also as the years go by our subjects get higher in the sky for us northerners. This will allow for fully correcting the dispersion.

 

Best Regards,

Russ



#11 Rustler46

Rustler46

    Viking 1

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

Posted 24 August 2019 - 12:40 AM

After dumping the 14 worst videos I have 13 videos that produce reasonable results. Each video was processed as follows in AutoStakkert!-2:

  • Best 5%, 12%, 30% and 75% of 15,000 frames, both sharpened (conv.tif) and unsharpened output (.tif)

As an experiment I've taken the 13 outputs (sharpened and unsharpened) in each percentage category and stacked these in AS!-2 as follows:

  • Best 4-, 6- 8- 13-frames (both sharpened and unsharpened stacked separately)

If I count correctly that gives 64 images that can be sent individually to Registax. I suspect the results will be different for the 32 AutoStakkert!-2 sharpened outputs versus the 32 produced from unsharpened outputs.

 

The sheer number of these to look at and process in Registax looks to be more than I will actually accomplish. My thinking is to take the stacks of unsharpened outputs and look at these first. Just a quick examination points to the stacks from the 5% and 12% video stacks looking more promising.

 

I don't know what this double stacking (stack of stacks) will do as for noise levels and resolution. Given that these videos were the best frames when seeing was only fair to good - I guess we will see how it turns out. 

 

Any comments or suggestions?

 

Best Regards,

Russ


Edited by Rustler46, 24 August 2019 - 02:05 AM.


#12 Tulloch

Tulloch

    Messenger

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

Posted 24 August 2019 - 01:09 AM

After dumping the 14 worst videos I have 13 videos that produce reasonable results. Each video was processed as follows in AutoStakkert!-2:

  • Best 5%, 12%, 30% and 75% of 1500 frames, both sharpened (conv.tif) and unsharpened output (.tif)

As an experiment I've taken the 13 outputs (sharpened and unsharpened) in each percentage category and stacked these in AS!-2 as follows:

  • Best 4-, 6- 8- 13-frames (both sharpened and unsharpened stacked separately)

If I count correctly that gives 64 images that can be sent individually to Registax. I suspect the results will be different for the 32 AutoStakkert!-2 sharpened outputs versus the 32 produced from unsharpened outputs.

 

The sheer number of these to look at and process in Registax looks to be more than I will actually accomplish. My thinking is to take the stacks of unsharpened outputs and look at these first. Just a quick examination points to the stacks from the 5% and 12% video stacks looking more promising.

 

I don't know what this double stacking (stack of stacks) will do as for noise levels and resolution. Given that these videos were the best frames when seeing was only fair to good - I guess we will see how it turns out. 

 

Any comments or suggestions?

 

Best Regards,

Russ

Hi Russ, just a few suggestions from me.

 

First, did you mean 1500 or 15000 frames? Hopefully the latter (more frames is better).

 

Second, before you process all those stacks in Registax, I would try a set with one video at 5%, 12%, 30% and 75%, find the best percentage and use that for all the other videos. 

 

Third, don't try to sharpen the sharpened outputs from AS!2 in Registax, it won't work. The conv files from AS!2 are meant to give a quick indication of which one will sharpen best in Registax.

 

Just my opinion on how I would tackle it, others may disagree ...

 

Andrew



#13 Rustler46

Rustler46

    Viking 1

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

Posted 24 August 2019 - 02:08 AM

Hi Russ, just a few suggestions from me.

 

First, did you mean 1500 or 15000 frames? Hopefully the latter (more frames is better).

 

Second, before you process all those stacks in Registax, I would try a set with one video at 5%, 12%, 30% and 75%, find the best percentage and use that for all the other videos. 

 

Third, don't try to sharpen the sharpened outputs from AS!2 in Registax, it won't work. The conv files from AS!2 are meant to give a quick indication of which one will sharpen best in Registax.

 

Just my opinion on how I would tackle it, others may disagree ...

 

Andrew

Yeah Andrew, I meant 15,000 frames per video.

 

I produced 5% 12% 30% and 75% for all the videos. Of course this just involved putting those four numbers in the 4 percentage boxes.

 

Oops! The photo shared in the original post of this thread is a Registax sharpened conv file. I'll give another go at using the unsharpened files.

 

It is easy to generate a lot of files with the software. But at least the tif output files are less than 2 MB each. When I get the best Saturn image from my latest imaging run, I'll be dumping the remaining 40 GB of videos and quite a few megabytes of tif images. After all I just want the best record of what Saturn looked like during that 2 hour imaging run.



#14 Kokatha man

Kokatha man

    Voyager 1

  • *****
  • Posts: 12833
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 24 August 2019 - 08:29 PM

I disposed of around 40 GB of videos by quickly analyzing them in AutoStakkert!-2. Starting with the earlier ones (when seeing was garbage) if there wasn't a significant fraction of frames above the 75% quality level - these were sent to the trash. Still more to cull out.

 

Russ

I haven't read all the posts so all of this might've been covered by others already Russ: don't reject captures merely because of "significant fractions above the 75% quality level" as we regularly go down to 60% in our stacks & quite often below this value.

 

Of course it also depends upon the quality of the best frame AS!3 makes it graph determination on & it is a good idea to examine the frame AS!3 determines is the "best" or "100% quality" frame...& the one where you plan to make your "cut-off" whatever % value that represents...no 2 captures are necessarily the same: a bad capture from start to finish might choose "the best of a bad bunch" for the #1 frame & that'll meant they're all lousy...& with 2 competing stacks where one has a very good "best frame" from "Analyse" & the other significantly poorer, you will have major outcome differences!

 

Re the ADC, the barlow's power is an important factor (with distances to the sensor affecting that figure, hence needing to be part of the equation, unless it is telenegative) - but I always inform folks of something John B. brought to my attention years ago now - the distance from ADC to sensor has a significant effect on the ability of it to correct for dispersion...the greater that distance is, the better the correction at lower elevations. wink.gif


  • Rustler46 likes this

#15 Rustler46

Rustler46

    Viking 1

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

Posted 24 August 2019 - 09:30 PM

I haven't read all the posts so all of this might've been covered by others already Russ: don't reject captures merely because of "significant fractions above the 75% quality level" as we regularly go down to 60% in our stacks & quite often below this value.

 

Of course it also depends upon the quality of the best frame AS!3 makes it graph determination on & it is a good idea to examine the frame AS!3 determines is the "best" or "100% quality" frame...& the one where you plan to make your "cut-off" whatever % value that represents...no 2 captures are necessarily the same: a bad capture from start to finish might choose "the best of a bad bunch" for the #1 frame & that'll meant they're all lousy...& with 2 competing stacks where one has a very good "best frame" from "Analyse" & the other significantly poorer, you will have major outcome differences!

 

Re the ADC, the barlow's power is an important factor (with distances to the sensor affecting that figure, hence needing to be part of the equation, unless it is telenegative) - but I always inform folks of something John B. brought to my attention years ago now - the distance from ADC to sensor has a significant effect on the ability of it to correct for dispersion...the greater that distance is, the better the correction at lower elevations. wink.gif

Thanks for the insight, Darryll - it's much appreciated. 

 

As for my rejected videos, they still reside in the trash folder. So If I run out of things to do with the better videos, these can be resurrected. Regarding the best frame in each video, will that be frame #1 in the new video after AutoStakkert!-2 has analyzed it? Watching that best to worst video play fills me with appreciation for the stacking process in pulling something useful out of the seeing mess. My location isn't noted for consistent good seeing, though I did note some excellent nights this past Spring. 

 

As for ADC to sensor spacing, placing the Barlow lens set on the camera nose piece does increase that spacing, at the expense of less amplification. But I'm happy with the f/17 it gives, particularly in my seeing regime. Even at that modest focal ratio, the sampling of 0.13 arc-seconds per pixel might be pushing things. The camera has small 2.9μ pixels. What do you think - is that over-sampled for a C-11? If so I could go with a 1.5X Barlow.

 

Does my plan to stack some of the tiffs resulting from stacking in AutoStakkert!-2 have merit? I've notice the quality graph of 13 frames shows a definite difference in their quality, decreasing down to around 50% at 10 frames. Also if I do stack these, should these only be from tiffs that were not sharpened in AutoStakkert!-2? Someone on the forum suggested not processing sharpened outputs.

 

I hope by the end of this coming week to decide on the best two or three images from my imaging run on August 20th. Really with observation lasting only 2 hours, I don't need 87 GB of videos left over. Just a few megabytes of the best images will suffice.

 

Best Regards,

Russ



#16 Rustler46

Rustler46

    Viking 1

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

Posted 25 August 2019 - 03:14 AM

Here's my latest attempt at getting the best result from the 2 hour Saturn imaging run of 2019 August 20.

 

Of the 27 videos (15,000 frames each) I kept the best 13 and stacked each of these at the 5%, 12%, 30% and 75% levels. Outputs from each of these were produced in both sharpened and unsharpened form. The outputs from each percentage group were stacked as follows:

  • For the 5% group the best 4, 6, 8 and all 13 were stacked again AutoStakkert!-2
  • For the 12% group the best 4, 6, 8 and all 13 were stacked again AutoStakkert!-2

This was done for both sharpened (conv) and unsharpened outputs. The results were quite marked when further sharpened in Registax.

  • First the stacked outputs from the 12% stacks were superior to those for the 5% stacks. The total frames stacked in the 12% group were 7200, 10,800, 14,400 and 23,400 frames.
  • For the lesser quality 5% stacks the total frame count was considerably lower - 3000, 4500, 6000 and 9750 frames.
  • Unexpectedly the stacks of sharpened outputs were superior to those that had not been sharpened by AutoStakkert!-2. While some of this could be due to my skills in using Registax, the unsharpened stacks just could not be made to equal the AS!-2 sharpened ones.
  • Since the AS!-2 sharpened 12% stacks were superior to the 5% ones, I'll need to do the same thing with the 30% stacks to see how they compare with the 12% stacks. While I'm tempted to just use the sharpened outputs, I give another try at the unsharpened ones too.

In any case this is the best I've been able to produce so far:

 

Screen Shot 2019-08-25 at 12.57.50 AM.png

C-1 f/17, ASI290MC, ADC, 14,400 frames stacked

 

If this ends up being my final "best image" for 2019 August 20, I'll need to do some touchup in Photoshop where the rings go behind the disc. I was pleased to have captured 5 belts and the polar zone, despite the so-so seeing and low elevation.

 

Comments and suggestions?

 

Regards,

Russ


Edited by Rustler46, 25 August 2019 - 03:40 AM.

  • Kenny V., Peter L. and happylimpet like this

#17 Rustler46

Rustler46

    Viking 1

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

Posted 26 August 2019 - 08:52 PM

Here's the best I could produce from the 30% stacks, with outputs stacked as described above:

 

Screen Shot 2019-08-26 at 5.33.45 PM.png

C-11, f/17, ASI290MC, ADC, 27,000 frames stacked

 

For comparison, here's the previous result of stacking the 12% outputs:

 

Screen Shot 2019-08-26 at 5.33.19 PM.png

C-11, f/17, ASI290MC, ADC, 14,400 frames stacked

 

I'll examine them more closely to decide which one to keep and touch up in Photoshop. Perhaps I'll keep both. Each has artifacts where the rings go behind the disc. I have already deleted 87 GB of videos and 1/2 GB of Tiff- and Png-format images. Only 14 images imported into Lightroom-5 were saved. Many of these will also be deleted. Time to move on.

 

Comments and suggestions?

 

Best Regards,

Russ


  • Kenny V. and happylimpet like this

#18 Kokatha man

Kokatha man

    Voyager 1

  • *****
  • Posts: 12833
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 26 August 2019 - 10:08 PM

Russ, the "best" frame after AS!3 analyses the avi is the "#1 frame" which is rated at "100%" & how the rest of the frames are compared to: be aware that this is based upon contrast (aka sharpness) & where you have the greatest contrast variation AS!3 determines said to be superior - which is why a split or broken frame rates so highly! :lol:

 

As to your 2nd query in Post #15: <"Does my plan to stack some of the tiffs resulting from stacking in AutoStakkert!-2 have merit? I've notice the quality graph of 13 frames shows a definite difference in their quality, decreasing down to around 50% at 10 frames. Also if I do stack these, should these only be from tiffs that were not sharpened in AutoStakkert!-2? Someone on the forum suggested not processing sharpened outputs.">

 

I'm not quite sure what you are referring to here...if these tifs are from different video captures then they will need to be integrated/derotated in WinJUPOS...so perhaps you'd best elaborate before I say something irrelevant. ;)

 

The Saturn images look a bit "washed" for colour - the ASI224MC, & I suspect the 290MC also suffers a bit there: if you have set the red & blue sliders to their correct values in FireCapture & applied the "RGB Balance" application in Registax6 they still invariably need some saturation & vibrancy boosting in post-processing...if you have any of the P/shop softwares try with a vibrancy of +23 & saturation of +10 for starters & take it from there...you can always experiment from that suggestion. (in Image> adjustments>vibrance...)


  • Kenny V. likes this

#19 Rustler46

Rustler46

    Viking 1

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

Posted 27 August 2019 - 12:51 AM

Russ, the "best" frame after AS!3 analyses the avi is the "#1 frame" which is rated at "100%" & how the rest of the frames are compared to: be aware that this is based upon contrast (aka sharpness) & where you have the greatest contrast variation AS!3 determines said to be superior - which is why a split or broken frame rates so highly! lol.gif

 

As to your 2nd query in Post #15: <"Does my plan to stack some of the tiffs resulting from stacking in AutoStakkert!-2 have merit? I've notice the quality graph of 13 frames shows a definite difference in their quality, decreasing down to around 50% at 10 frames. Also if I do stack these, should these only be from tiffs that were not sharpened in AutoStakkert!-2? Someone on the forum suggested not processing sharpened outputs.">

 

I'm not quite sure what you are referring to here...if these tifs are from different video captures then they will need to be integrated/derotated in WinJUPOS...so perhaps you'd best elaborate before I say something irrelevant. wink.gif

 

The Saturn images look a bit "washed" for colour - the ASI224MC, & I suspect the 290MC also suffers a bit there: if you have set the red & blue sliders to their correct values in FireCapture & applied the "RGB Balance" application in Registax6 they still invariably need some saturation & vibrancy boosting in post-processing...if you have any of the P/shop softwares try with a vibrancy of +23 & saturation of +10 for starters & take it from there...you can always experiment from that suggestion. (in Image> adjustments>vibrance...)

Thanks for the most useful feedback and tips, Daryll. I hope I spell your name correctly.

 

Yes, the 13 images I stacked are the resulting tiff files that AutoStakkert!-2 produced (best 30% of the video frames stacked) from the best 13 avi videos. So 13 videos with the best 30% stacked produced 13 tiff format images. What I've done is just feed these 13 tiff-format files into AS!-2 and stacked the best 4, 6, 8 and 13 frames. Basically I'm stacking the results of what had been stacked from avi videos. These videos were captured over a period of perhaps 1/2 hour.

 

I repeated the same process using the best 13 videos (stacked at the best 12%). The 13 tiff-format files were fed a second time into AutoStakkert!-2.

 

For me to get AutoStakkert!-3 working, I'll need to download the 64-bit version of Wine. Emil Kraaikamp clued me in here as to why I could not get AS!-3 working on my MacBook computer. My version of the Wine wrapper must be the 32-bit version.

 

I'm not familiar with WinJUPOS. But I'm hoping that a total time spread of 1/2 hour won't smear things too much. Perhaps that's wishful thinking. If I was using only a single video, the 2-1/2 minute duration should be OK. I'm just trying to extract some more detail from a night of mediocre seeing. The orientation of Saturn in each video was identical on the camera's sensor. There really doesn't appear to be much in the way of details on the disc apart from the 5 belts and polar zone. When using FireCapture 2.6, I just kept the default settings and captured un-deBayered RGB video in avi-format. Emil Kraaikamp recommends using the ser-format. So perhaps I'll give that a try. Avi format seems to work well.

 

Here are the two images after I applied your suggested vibrancy and saturation parameters in Lightroom-5. Then I did an auto white balance. I also reduced clarity down to 0 (no correction). I'm not sure what "clarity" is. But it's grouped with saturation and vibrance in Lightroom. It does apply some measure of sharpening. Reducing clarity also reduced some of the artifacts at the ring/disc juncture. This improvement was most noticeable on the 30% image (right side image below). Here's what resulted:

 

Screen Shot 2019-08-26 at 9.56.28 PM.png

The 12% videos stack on on left, 30% on the right.

 

These two weren't processed identically in Registax. They were just sharpened the best I could without too much noise or ring/disc artifacts. To my eye the left (12%) image looks best. Perhaps it is a bit over sharpened. The 30% image just looked blurred with details like ring division and belts hard to pull out in Registax. Anyway I'm going to do some "artistic adjustments" in Photoshop on both images. 

 

I hope to have another opportunity to capture more videos of Jupiter and Saturn this year. But they are getting past best viewing. Saturn might be the best subject for the remainder of the season.

 

Best Regards,

Russ


Edited by Rustler46, 27 August 2019 - 02:02 AM.


#20 Kokatha man

Kokatha man

    Voyager 1

  • *****
  • Posts: 12833
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 27 August 2019 - 07:24 PM

Clarity is probably some form of sharpening Russ...opening RAW images from my dslr in CS4 has a similar application.

 

Yes, you will have to learn how to use WinJUPOS to combine those tifs...but for a single video capture on Saturn you can go for much longer than 2.5 minutes, at the f/l's you've employed: 6 minutes is absolutely fine no matter what anyone says, & I've never seen any difference at all between 8bit avi data & ser captures - speaking from the only considerations we apply - the end result! ;)


  • Rustler46 likes this

#21 Rustler46

Rustler46

    Viking 1

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

Posted 29 August 2019 - 12:34 AM

Thanks for the most useful feedback and tips, Daryll. I hope I spell your name correctly.

I see from your website it's Darryl. I hope I remember.



#22 Rustler46

Rustler46

    Viking 1

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

Posted 31 August 2019 - 12:45 AM

Here's my final iteration of the image of Saturn on 2019 August 20. I did a few "artistic adjustments" in Photoshop Elements 2.0 to get rid of some stacking artifacts. While it's much improved, there are still a few warts visible on close inspection. But it's the best I can do for now, especially considering my still developing skills, so-so seeing and Saturn's low elevation. 

 

Screen Shot 2019-08-30 at 10.34.03 PM.png

Celestron-11, f/17, ZWO ASI290MC camera, ZWO ADC, elevation 23°

 

Video Capture:

FireCapture 2.6 on MacBook Pro (mid-2012), 16GB RAM, 500 GB SSD, USB 3.0, Celestron-11 collimation was near perfect.

 

Processing:

13 videos of 15,000 frames each (~ 2-1/2 minutes), best 12% of frames in each video stacked, best 8 output tiffs stacked, 14,400 frames in total, Registax-6 wavelet sharpened, Photoshop Elements 2.0 touchup, final adjustments in Lightroom-5

 

One point I've come to appreciate is to establish a tight region of interest (ROI) around Saturn when capturing the videos. Then using the auto-track feature of FireCapture 2.6, the file size is kept within reasonable limits for whatever length video is attempted. Also I'll need to install AutoStakkert!-3 (64-bit) to get around the 4 GB file size limit with AS!-2 that I'm presently using. Even when using auto-track during video capture, it is necessary to do some manual guiding to keep the planet within the ROI. It would be good if I enabled periodic error correction on my Losmandy G-11 mount to limit tracking errors. But auto-track does a decent job of keeping the planet centered in each frame. I just need to keep monitoring the process to apply corrections when necessary. There is much to learn in this craft. But progress has been made in that regard.

 

So it's time to move on. I'll be deleting all videos, intermediate images, inferior stacks and so on, just saving the best results. Even with 250 GB free space on my MacBook's SSD, it is easy to quickly use up a lot of space. For example on August 20th, I gathered 87 GB in 2-1/2 hours. I'm slowly learning how best to practice lunar/planetary imaging - "dusting off the cobwebs" so to speak. Thanks for the help from forum members. It is much appreciated.

 

Comments & suggestions?

 

Best Regards,

Russ


Edited by Rustler46, 01 September 2019 - 12:41 AM.



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