After the taster, here's the main course.... Jupiter 27-28 Oct 2023

[geoflewis]

Following my earlier post....

....here is the main course, well most of it anyway. I've processed 5 sets of 20 SERs through derotation and the rest, so will let the pictures speak for them selves.

Not sure why the 3rd image has turned out a bit bluer than the others, but it is also the sharpest, so maybe that was from when I had the clearest sky and best transparancy. The processing for all images was almost identical.

I still have 2 set to process, which I will add later, but I'm totally whacked and offered to show a friend my processing techniques tomorrow so will leave them for that.

Thanks for looking.

Edited October 29, 2023 by geoflewis
added final 2 images from the session

[Barv]

excellent work Geof!

[Kon]

Fantastic images! First and second to last are my favourite. IMHO, the third one could do with a bit less sharpening.

[Space Cowboy]

Plenty of detail on those Geof! The last 2 images are very good. 👍

[Pete Presland]

Terrific set of images, a fine level of details visible in all of them.

[geoflewis]

2 hours ago, Kon said:

Fantastic images! First and second to last are my favourite. IMHO, the third one could do with a bit less sharpening.

2 hours ago, Space Cowboy said:

Plenty of detail on those Geof! The last 2 images are very good. 👍

2 hours ago, Pete Presland said:

Terrific set of images, a fine level of details visible in all of them.

Thanks guys.

@Kon, it's interesting isn't it, as I applied identical sharpening to all the images, but I have to agree with you that it looks too much on the 3rd one. I'll probably revisit it and see if I can dial it back a tad. Thanks for the helpful critique.

 

 

[geoflewis]

If anyone is interested I processed the last 2 sets of data from the other night and added them alonside the others at the start of this thread. The final image has less data, mostly because my hard drive was full, but in any case the sky was deteriating quickly with thickening cloud and I'd only kept going in the hope that I could capture the completion of Europa's transit. It's shadow had exited right, but Europa still had a few minutes still to go.

My next challeng will be to try to generate a GIF of the entire session, but that's probably for another day. I have just checked Jupiter's Moons and see that the GRS will be transiting tonight between 10pm and 3am without any moons, but do I want another long session; I'm still recovering from the previous two.... 🤔

[Kon]

3 hours ago, geoflewis said:

Thanks guys.

@Kon, it's interesting isn't it, as I applied identical sharpening to all the images, but I have to agree with you that it looks too much on the 3rd one. I'll probably revisit it and see if I can dial it back a tad. Thanks for the helpful critique.

 

 

I wonder if the signal was a bit weaker on this capture and therefore the sharpening was a bit too strong.

[Kon]

45 minutes ago, geoflewis said:

If anyone is interested I processed the last 2 sets of data from the other night and added them alonside the others at the start of this thread. The final image has less data, mostly because my hard drive was full, but in any case the sky was deteriating quickly with thickening cloud and I'd only kept going in the hope that I could capture the completion of Europa's transit. It's shadow had exited right, but Europa still had a few minutes still to go.

My next challeng will be to try to generate a GIF of the entire session, but that's probably for another day. I have just checked Jupiter's Moons and see that the GRS will be transiting tonight between 10pm and 3am without any moons, but do I want another long session; I'm still recovering from the previous two.... 🤔

Yes you do! Think of all the nights that you have missed because of bad weather.

[geoflewis]

8 minutes ago, Kon said:

I wonder if the signal was a bit weaker on this capture and therefore the sharpening was a bit too strong.

I took another look at that data set and IIRC I was still experimenting with the best way to handle the shadow transit, so had a few different versions. I've just reprocessed it and can see that I did in fact increase the HF wavelet strength in my original, so have backed that off to where I processed the others. It's subtle but I thin better in version 2, so what do you think?

Edited October 29, 2023 by geoflewis

[Mr Spock]

Very nicely done. Lots of detail there 👍

[geoflewis]

7 minutes ago, Mr Spock said:

Very nicely done. Lots of detail there 👍

Thanks Michael

[Kon]

23 minutes ago, geoflewis said:

I took another look at that data set and IIRC I was still experimenting with the best way to handle the shadow transit, so had a few different versions. I've just reprocessed it and can see that I did in fact increase the HF wavelet strength in my original, so have backed that off to where I processed the others. It's subtle but I thin better in version 2, so what do you think?

 

Much much nicer. Still loads of details but more subtle. Of all the captures I think that's the best one now, in terms of details 😉 (nice processing on all).

[geoflewis]

1 minute ago, Kon said:

Much much nicer. Still loads of details but more subtle. Of all the captures I think that's the best one now, in terms of details 😉 (nice processing on all).

Thanks Kostas, I did think that was the best set, so of course I had to ruin it by pushing too hard. There was so much processing to do that I was getting cross-eyed or screen blindness. I had an astronomy friend over today who wanted to watch me process some data, so I slowly took him through Set 6 of 7, then later completed the final set; it took me about 3 hours to demo whilst also explaining the purpose of each step.

[Kon]

3 minutes ago, geoflewis said:

Thanks Kostas, I did think that was the best set, so of course I had to ruin it by pushing too hard. There was so much processing to do that I was getting cross-eyed or screen blindness. I had an astronomy friend over today who wanted to watch me process some data, so I slowly took him through Set 6 of 7, then later completed the final set; it took me about 3 hours to demo whilst also explaining the purpose of each step.

Don't we always? I do the same and I always appreciate when others spot it. I am always open to constructive criticism.

I am sure your friend picked some excellent processing skills from you.

[geoflewis]

Just now, Kon said:

Don't we always? I do the same and I always appreciate when others spot it. I am always open to constructive criticism.

I am sure your friend picked some excellent processing skills from you.

....and I equally appreciate your's and others critique.

My friend is strictly a DSO imager, mostly with a short tube refactor, but has always wanted to know what I do to get my planetary images. It was great showing the steps involved to someone that understands the challenges of astrophotography, but has never dabbled with planetary imaging.

[geoflewis]

Having completed processing the 7 sets of data to get high resolution images, I thought this might be a good way to show the progress of Jupiter and Io through the night of 27/28 Oct 2023. You can zoom in and scroll through them side by side without having to jump between different images. I still have the GIF to consider, but I won't be able to show that in such high resolution, as it's too much work fixing the smeared Europa and shadow on the large quantity of derotated stacks that I need for a smooth GIF presentation, so I'll probably be using the individally processed 'best 2000 frame' stacks from the 133 videos that I captured.

 

[knobby]

Massive effort and well worth it ! really nice.

[orion25]

Brilliant, Goef! Very nice set!

 

Regards,

Reggie🤩

[geoflewis]

6 minutes ago, knobby said:

Massive effort and well worth it ! really nice.

3 minutes ago, orion25 said:

Brilliant, Goef! Very nice set!

 

Regards,

Reggie🤩

Thanks both, yes a lot of work and a huge amout of data. I keep all my data both on local HDs and also upload to cloud storage. This session was over 330GB of data in 133 SERs and associated files. I started to upload it to the cloud yesterday afternoon, and currently some 20 hours later it's still esimating 8 more hours to complete the upload. Unfortunately living in a rural location, I don't have the possibility of FTTH (very fast fibre), so it's a long slow process.

 

[Space Cowboy]

Wow, you've been busy Geof! Those repros look very nice! 

I take it video derotation doesn't mess up the moon transit like derotation of images?  I thought maybe that problem had been fixed but Derot of images messed up mine.

I use a HDD rather than cloud.... enough of the bloody stuff around here already 😜

[geoflewis]

2 hours ago, Space Cowboy said:

Wow, you've been busy Geof! Those repros look very nice! 

I take it video derotation doesn't mess up the moon transit like derotation of images?  I thought maybe that problem had been fixed but Derot of images messed up mine.

I use a HDD rather than cloud.... enough of the bloody stuff around here already 😜

Thanks Stuart,

I'm not derotating the videos, but derotating the TIFF stack of best 2000 frames from each video. Unfortunately the moon correction tool in WinJupos doesn't do a great job, when in actual transit, but it's just about ok when the moon is still off to the side.. It does align the moon and shadow, but leaves a nasty artifact around them, so I still have to photoshop (I actually use Affinity Photo now, but you know what I mean) the moon and shadow back in from a lower resolution single video stack, with as near identical time stamp as the derotate stack. To be honest it's a complete faff, but when the data is good its worth the extra messing around. However, it's why I'm not going to go for high res data if I do compile the GIF - even I don't have the stomach for photoshop moon and shadow corrections on ~100 separate images.

I do have several external HDs for my archive, but a couple of years ago when I went to retrieve something, one of them the drive had died, losing several years of data. Actually I was fortunate that I had a lot of that on more than one drive, but I couldn't recover that HD I sent it to a friend who works in the IT industry, who was able to recover the images (SERS and TIFFs), but all the associted EXIF data, etc., was lost, so I just had a buch of videos with no file names, date, or time stamp, so pretty damn useless to me. After that I took the decision to upload to cloud as well as keeping local HD storage - so belt and braces.

But you're right, there's definitely too much cloud.....🤪

[Space Cowboy]

15 hours ago, geoflewis said:

Thanks Stuart,

I'm not derotating the videos, but derotating the TIFF stack of best 2000 frames from each video. Unfortunately the moon correction tool in WinJupos doesn't do a great job, when in actual transit, but it's just about ok when the moon is still off to the side.. It does align the moon and shadow, but leaves a nasty artifact around them, so I still have to photoshop (I actually use Affinity Photo now, but you know what I mean) the moon and shadow back in from a lower resolution single video stack, with as near identical time stamp as the derotate stack. To be honest it's a complete faff, but when the data is good its worth the extra messing around. However, it's why I'm not going to go for high res data if I do compile the GIF - even I don't have the stomach for photoshop moon and shadow corrections on ~100 separate images.

I do have several external HDs for my archive, but a couple of years ago when I went to retrieve something, one of them the drive had died, losing several years of data. Actually I was fortunate that I had a lot of that on more than one drive, but I couldn't recover that HD I sent it to a friend who works in the IT industry, who was able to recover the images (SERS and TIFFs), but all the associted EXIF data, etc., was lost, so I just had a buch of videos with no file names, date, or time stamp, so pretty damn useless to me. After that I took the decision to upload to cloud as well as keeping local HD storage - so belt and braces.

But you're right, there's definitely too much cloud.....🤪

Thanks for the feedback Geof. I might try video derotation  just to compare . It must be 10 years since I tried derotation on a transit.

I can see why you use the cloud. At least they come in useful for something 😉

[geoflewis]

31 minutes ago, Space Cowboy said:

I might try video derotation  just to compare .

I'm not sure how video derotation would help unless you were captureing everything in a very long SER file, as AS3! has no issue aligning the moon and shadows from a 1m SER.

When I tried video derotation of Mars last year the file size of the derotated SER was approximately double the original SER, i.e a 6min SER at 3ms (330fps) with the small ROI 300x300 required for Mars was approx 10 GB, which after derotation was ~20GB. When I tried a larger ROI of 400x400 the original 6m SER was 18GB and the derotated SER 36GB. This was far too large, so I gave up derotating the SERs for Mars, as the rotation was minimal in 6 mins and I couldn't tell the difference.

If I was going to try SER derotation for Jupiter, then I guess my workflow might go something like this:

1. Capture a very long video, say 10m-15m (based on my 8ms capture speed, with the much larger ROI (600x600), the initial file would be ~50GB...🤯)
2. Grade the video in PIPP to keep, say, just the best 20%, which I think would reduce the file to ~10GB
3. Derotate the graded video in WinJupos.
4. Stack in AS3!
5. Other post processing, in Registax, Astrosurface, Image Analyser, etc.

I'll be iterested to learn what you find if you do try it.

 

[Space Cowboy]

57 minutes ago, geoflewis said:

I'm not sure how video derotation would help unless you were captureing everything in a very long SER file, as AS3! has no issue aligning the moon and shadows from a 1m SER.

When I tried video derotation of Mars last year the file size of the derotated SER was approximately double the original SER, i.e a 6min SER at 3ms (330fps) with the small ROI 300x300 required for Mars was approx 10 GB, which after derotation was ~20GB. When I tried a larger ROI of 400x400 the original 6m SER was 18GB and the derotated SER 36GB. This was far too large, so I gave up derotating the SERs for Mars, as the rotation was minimal in 6 mins and I couldn't tell the difference.

If I was going to try SER derotation for Jupiter, then I guess my workflow might go something like this:

1. Capture a very long video, say 10m-15m (based on my 8ms capture speed, with the much larger ROI (600x600), the initial file would be ~50GB...🤯)
2. Grade the video in PIPP to keep, say, just the best 20%, which I think would reduce the file to ~10GB
3. Derotate the graded video in WinJupos.
4. Stack in AS3!
5. Other post processing, in Registax, Astrosurface, Image Analyser, etc.

I'll be iterested to learn what you find if you do try it.

 

Yep I used vid Derot on Mars a couple of times. The key for me is the field derotation which Derot of images does not do so possibly that would help with the transit.

I've a feeling someone on CN said vid Derot was better for transits or I could have dreamt it. 🙂

Edited October 31, 2023 by Space Cowboy