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

Jupiter with GRS Time Lapse 2024-10-07

  • Please log in to reply
19 replies to this topic

#1 barrett_flansburg

barrett_flansburg

    Ranger 4

  • -----
  • topic starter
  • Posts: 374
  • Joined: 30 Dec 2014
  • Loc: New Mexico Astronomy Village

Posted 07 October 2024 - 01:55 PM

This is a time lapse video of Jupiter and its Great Red Spot taken the morning of October 7, 2024 from my backyard in the New Mexico Astronomy Village near Deming, New Mexico. It was supposed to be about three times as long with the GRS swinging all the way from one side of the planet to the other, but my dome disconnected in the middle  of capturing the images so most of the event was lost. It was taken through a Stellarvue SVX180T refractor telescope with a 2X Barlow lens and a ZWO ASI715MC camera. 82 SER videos were recorded using FireCapture between 0911 and 1039 UT. These were stacked using AutoStakkert 4, sharpened and color balanced in WaveSharp 1.0 Beta, with final processing in Photoshop. WaveSharp was used to create the animated PNG file.

 

This is my first attempt to link to an animated PNG file from my gallery, so hopefully it will work!

 

gallery_237968_24628_10207031.png

 


  • Skywatchr, elwaine, sfugardi and 10 others like this

#2 KiwiRay

KiwiRay

    Gemini

  • *****
  • Posts: 3,271
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 07 October 2024 - 02:23 PM

Good work making such a smooth animation. Colours look a bit odd, with a strong purple tint. I didn't know you could have an apng animation play within a post when linked from your gallery. Other people's posts required clicking on a static image to open a new window.


Edited by KiwiRay, 07 October 2024 - 04:03 PM.


#3 Tulloch

Tulloch

    Fly Me to the Moon

  • *****
  • Posts: 7,229
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 07 October 2024 - 03:29 PM

Other people's posts required clicking on a static image to open a new window.

Sometimes it works, sometimes it doesn't. I think it has something to do with the physical size of the image, but I'm not sure.


  • KiwiRay likes this

#4 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 07 October 2024 - 03:51 PM

Good work making such a smooth animation. Colours looks a bit odd, with a strong purple tint. I didn't know you could have an apng animation play within a post when linked from your gallery. Other people's posts required clicking on a static image to open a new window.

It works fine. The problem is it probably violates the TOS because it is inlined and probably > 500 KB.



#5 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 07 October 2024 - 03:53 PM

But, great animation.

 

This is probably the best example I've seen of why I think the most appropriate approach to "derinding" an image is to simply mask out the outer ring(ing). It is very clear that is a ghost caused by the first diffraction ring that is larger than the proper size of the disk.


Edited by Borodog, 07 October 2024 - 06:03 PM.


#6 barrett_flansburg

barrett_flansburg

    Ranger 4

  • -----
  • topic starter
  • Posts: 374
  • Joined: 30 Dec 2014
  • Loc: New Mexico Astronomy Village

Posted 07 October 2024 - 05:22 PM

Mike,

 

Would I draw the mask by hand for one still image from the time lapse (since Jupiter isn't a circle), then mask out the rind for all the images using the same mask? And finally create the time lapse from the edited still images.


  • Borodog likes this

#7 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 07 October 2024 - 06:03 PM

Mike,

 

Would I draw the mask by hand for one still image from the time lapse (since Jupiter isn't a circle), then mask out the rind for all the images using the same mask? And finally create the time lapse from the edited still images.

That is exactly what I do.



#8 barrett_flansburg

barrett_flansburg

    Ranger 4

  • -----
  • topic starter
  • Posts: 374
  • Joined: 30 Dec 2014
  • Loc: New Mexico Astronomy Village

Posted 08 October 2024 - 02:32 PM

Mike,

 

Here is my attempt at de-rinding the Jupiter video. I didn't use masks. Instead, here are my steps in Photoshop:

 

1. Create a new action "De-Rind".

2. Select Object to select Jupiter including the rind.

3. Use arrow keys to move the selection edge from the rind edge to the edge of Jupiter.

4. Feather the selection by 2 pixels to make it less artificial-looking.

5. Invert the selection.

6. Edit/Clear the selection to black.

7. Save the edited image to a folder for the de-rind images.

 

I then applied this action to all of the previous images and made a new time lapse video of the result. Looks much better to me! What do you think?

 

https://www.cloudyni...28_11709803.png



#9 Winteria

Winteria

    Viking 1

  • -----
  • Posts: 528
  • Joined: 23 Aug 2022
  • Loc: Seattle, USA

Posted 08 October 2024 - 07:00 PM

Mike,

 

Here is my attempt at de-rinding the Jupiter video. I didn't use masks. Instead, here are my steps in Photoshop:

 

1. Create a new action "De-Rind".

2. Select Object to select Jupiter including the rind.

3. Use arrow keys to move the selection edge from the rind edge to the edge of Jupiter.

4. Feather the selection by 2 pixels to make it less artificial-looking.

5. Invert the selection.

6. Edit/Clear the selection to black.

7. Save the edited image to a folder for the de-rind images.

 

I then applied this action to all of the previous images and made a new time lapse video of the result. Looks much better to me! What do you think?

 

https://www.cloudyni...28_11709803.png

Well you've certainly removed the rind!

 

Personally I don't like that you've essentially erased the entire left limb from the image.


  • KiwiRay likes this

#10 KiwiRay

KiwiRay

    Gemini

  • *****
  • Posts: 3,271
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 08 October 2024 - 08:19 PM

Well you've certainly removed the rind!

 

Personally I don't like that you've essentially erased the entire left limb from the image.

Yeah, the rind edge is the edge of the planet - the rind is not an extra bit stuck onto the side. It's not the first time I've seen that done (I think someone else used that approach in an image posted last week), but it's a pretty brutal way of dealing with it.


Edited by KiwiRay, 08 October 2024 - 10:45 PM.

  • Winteria likes this

#11 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 08 October 2024 - 08:42 PM

Mike,

Here is my attempt at de-rinding the Jupiter video. I didn't use masks. Instead, here are my steps in Photoshop:

1. Create a new action "De-Rind".
2. Select Object to select Jupiter including the rind.
3. Use arrow keys to move the selection edge from the rind edge to the edge of Jupiter.
4. Feather the selection by 2 pixels to make it less artificial-looking.
5. Invert the selection.
6. Edit/Clear the selection to black.
7. Save the edited image to a folder for the de-rind images.

I then applied this action to all of the previous images and made a new time lapse video of the result. Looks much better to me! What do you think?

https://www.cloudyni...28_11709803.png



Looks fine to me. Yes, you are losing the “true” edge (I proved this to my own satisfaction today), but it looks much better.

#12 Winteria

Winteria

    Viking 1

  • -----
  • Posts: 528
  • Joined: 23 Aug 2022
  • Loc: Seattle, USA

Posted 08 October 2024 - 08:44 PM

Barry - The new version of WaveSharp that’s currently in testing has a “de-rind” feature that seems to work pretty well and doesn’t cut off bits of the disk. Whenever that version is released it should be a much better solution for those who dislike the rind.



#13 barrett_flansburg

barrett_flansburg

    Ranger 4

  • -----
  • topic starter
  • Posts: 374
  • Joined: 30 Dec 2014
  • Loc: New Mexico Astronomy Village

Posted 08 October 2024 - 09:38 PM

Well you've certainly removed the rind!

Personally I don't like that you've essentially erased the entire left limb from the image.

 

Yeah, the rind edge is the edge of the Planet - the rind is not an extra bit stuck onto the side. It's not the first time I've seen that done (I think someone else used that approach in an image posted last week), but it's a pretty brutal way of dealing with it.

 

Looks fine to me. Yes, you are losing the “true” edge (I proved this to my own satisfaction today), but it looks much better.

So are you guys saying that the rind is an image of a real edge of the planet, not just a superfluous ghost image that duplicates the real edge of the planet? Hmmm… it certainly looks better without the rind. And it doesn’t look asymmetric to me with the rind removed. 
 

I will be interested to see how the rind tool in WaveSharp works. I know Don has been using it and has positive things to say about it. 



#14 Winteria

Winteria

    Viking 1

  • -----
  • Posts: 528
  • Joined: 23 Aug 2022
  • Loc: Seattle, USA

Posted 08 October 2024 - 11:35 PM

So are you guys saying that the rind is an image of a real edge of the planet, not just a superfluous ghost image that duplicates the real edge of the planet? Hmmm… it certainly looks better without the rind. And it doesn’t look asymmetric to me with the rind removed. 

Yes, what we call the 'rind' is a real part of the planet, separated from the rest of it by a dark division just inside the limb which is caused by the gibbs effect.

 

To each their own! I don't think it looks terrible either, I just like to present my images as faithfully to the data as possible.



#15 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 09 October 2024 - 10:31 AM

Yes, what we call the 'rind' is a real part of the planet, separated from the rest of it by a dark division just inside the limb which is caused by the gibbs effect.

 

To each their own! I don't think it looks terrible either, I just like to present my images as faithfully to the data as possible.

Then you would not do anything about the rind. :O)

 

"The rind" is the ringing you get because the finite aperture has filtered out the high frequencies required to produce a sharp edge. Here's a simulated rind I made by taking an under sampled (i.e. sharp) image of Jupiter and filtering out the high frequencies in waveSharp. The diameter of the disk does not change.

 

image(9).png

 

Yesterday I was thinking "the rind" corresponded to the first diffraction ring, which would be outside the proper diameter of the planet, but that is definitely wrong (for one thing it is too bring; the ring around the planet associated with the first diffraction ring is in fact there, and it is outside the proper diameter of the planet, but it is faint, as you would expect).



#16 Kokatha man

Kokatha man

    James Webb Space Telescope

  • *****
  • Posts: 18,339
  • Joined: 13 Sep 2009
  • Loc: "cooker-ta man" downunda...

Posted 09 October 2024 - 06:44 PM

In nearly all examples where we see a so-called "rind" on Jupiter we are actually looking at "ringing" aka the "Gibbs Effect"..! My "nearly all" refers to the root cause of "ringing" also - pushing/sharpening data to said point! ;)

 

I'm fine with whatever folks think is ok in processing - "each to their own" as my old Mum used to say...but chopping off some of the disk is "beyond the pale" imo grin.gif  - and something ALPO & PVOL etc would certainly not want to see if folks are thinking of submitting their images! shocked.gif

 

In some ways similar to part of how I deal with the actual diffraction effects on Mars (the real "rind" imo) over-sharpening's ringing on Jove can easily be dealt with by the use of a lesser-sharpened image's limb, used as a mask in P/shop etc.....I have some sympathy for doing this as image outcomes can be a bit frustrating if the seeing is not cooperating - and there is often a tendency to over-sharpen lesser data in processing! ;) 

 

Any other effects outside of the limbs are easily dealt with by adjusting the black-point in "levels" adjustments btw. (which usually requires some adjustment of the other 2 sliders)

 

For anyone interested in the NDD&B approach to the diffraction artefact on Mars there are a couple of tutes of mine on CN - unfortunately the earlier, more extended one with "layer masks" is somewhere in the archives, but the one showing the NDD&B aspects is readily accessible - I dug it out for Don in the last week. 



#17 KiwiRay

KiwiRay

    Gemini

  • *****
  • Posts: 3,271
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 09 October 2024 - 10:32 PM

Then you would not do anything about the rind. :O)

 

"The rind" is the ringing you get because the finite aperture has filtered out the high frequencies required to produce a sharp edge. Here's a simulated rind I made by taking an under sampled (i.e. sharp) image of Jupiter and filtering out the high frequencies in waveSharp. The diameter of the disk does not change.

 

attachicon.gif image(9).png

 

Yesterday I was thinking "the rind" corresponded to the first diffraction ring, which would be outside the proper diameter of the planet, but that is definitely wrong (for one thing it is too bring; the ring around the planet associated with the first diffraction ring is in fact there, and it is outside the proper diameter of the planet, but it is faint, as you would expect).

I know nothing of the physics of this, but the dark inside rind artifact is the same distance from the edge as the bright outside ring. Are you saying that they are not generated by the same process with one being an exterior diffraction ring and the other a corresponding interior diffraction ring?


Edited by KiwiRay, 09 October 2024 - 10:33 PM.


#18 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 10 October 2024 - 11:56 AM

I know nothing of the physics of this, but the dark inside rind artifact is the same distance from the edge as the bright outside ring. Are you saying that they are not generated by the same process with one being an exterior diffraction ring and the other a corresponding interior diffraction ring?

It's complicated. You can think of the image as though it is an underlying sharp (under sampled) "true" image that has been convolved with the Airy disk. But that isn't quite right. If that were the case, you could at least in principle deconvolve the Airy disk back out if the SNR were high enough and you had a perfect knowledge of the true Airy disk, ending up with a perfectly sharp (under sampled) image again. But you can't, because that isn't what has happened. What has happened is that the finite aperture has filtered out the higher frequencies required to make a perfectly sharp image. You can't deconvolve them back out because they were not recorded. You can in principle deconvolve atmospheric turbulence if the SNR is high enough and you have a good enough estimate of the point spread function. So when we sharpen we compensate for atmospheric turbulence, but we cannot really compensate for diffraction effects. Hence when we image stars and sharpen them, we get sharp images of Airy disks, not single pixels containing all the light. If you try to do that, all you can end up with is noise, since that is the only thing that has reached the sensor at those frequencies.

 

Diffraction effects are the result of high frequency filtering, and the ringing that results from not having the high frequencies required to make perfectly sharp edges (or points, in the case of the Airy disk), like this "image" of a perfectly sharp square wave (like a cross section of intensity on a line cutting through Mars, for example), that is formed by having successively higher frequencies added or removed:

 

square.png

 

You can see that near the sharp edge the intensity actually over shoots above that of the "true" image. And just inside that it undershoots. That's the bright and dark rings of the rind. Sometimes you will see further bright and dark fringes of decreasing amplitude as you move in, but they are usually lost in the features and non-uniformity of the planet's intensity.

 

It's a mistake to blame this all on over-sharpening. It's a real optical effect that may be harder to see under some circumstances. It's harder to see in Jupiter than Mars because Jupiter has a softer edge. You don't see this effect on the terminator of Mars because that is essentially a very soft edge. When I say "soft edge" I mean that there are fewer, if any, higher frequencies required to make the image of that edge, and hence the high frequency filtering has less, if any, effect there.

 

This example is not exact, by the way; this is a truncated Fourier series on a finite domain. The image plane is not constrained in the same way. In this example adding higher frequencies brings the image closer to the true diameter of the true image. But in the real world the ringing is not constrained in this way, and the last bright fringe really does more or less correspond to the position of the true limb of the planet.

 

There is a faint ring outside of the planet (perhaps even two or three) that "corresponds" to the diffraction rings in a more or less straightforward way.


  • theaberrator and KiwiRay like this

#19 barrett_flansburg

barrett_flansburg

    Ranger 4

  • -----
  • topic starter
  • Posts: 374
  • Joined: 30 Dec 2014
  • Loc: New Mexico Astronomy Village

Posted 10 October 2024 - 02:10 PM

Mike,

 

Thank you for an excellent explanation. It sounds like it is all my fault for using a finite aperture! lol.gif Actually, I'm sure the small 7-inch aperture does exaggerate the problem, but it is much less sensitive to seeing than my C14 so it is a worthwhile tradeoff.

 

Just to put things into perspective, I created a video showing only the rind that I removed from the images. I expected to be able to make out the higher-contrast bands, but I can't. In my opinion, this demonstrates that I didn't lose anything by cutting out the rind. But it does look like a cool video of a crescent Jupiter as seen from a spacecraft beyond its orbit...

 

https://www.cloudyni...628_5892788.png


  • Borodog likes this

#20 Borodog

Borodog

    Voyager 1

  • *****
  • Posts: 11,801
  • Joined: 26 Oct 2020
  • Loc: St. Augustine, FL

Posted 11 October 2024 - 02:29 PM

Mike,

 

Thank you for an excellent explanation. It sounds like it is all my fault for using a finite aperture! lol.gif Actually, I'm sure the small 7-inch aperture does exaggerate the problem, but it is much less sensitive to seeing than my C14 so it is a worthwhile tradeoff.

 

Just to put things into perspective, I created a video showing only the rind that I removed from the images. I expected to be able to make out the higher-contrast bands, but I can't. In my opinion, this demonstrates that I didn't lose anything by cutting out the rind. But it does look like a cool video of a crescent Jupiter as seen from a spacecraft beyond its orbit...

 

https://www.cloudyni...628_5892788.png

Yes. I still think this is a perfectly valid way to treat the rind in order to get an aesthetically pleasing result.




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