Saturn June 2nd Good seeing at last!

[U.K.Cowboy]

Good seeing at last has produced my best Saturn images of the year. 2 images below using 1.5x drizzle, 1st is a single avi taken at 21.48 UT and 2nd shot is 20 mins of captures de-rotated in winjup.
The 1st image has a sharper hex but the de-rot image is super smooth. Anyhow not bad for 25 degrees elev!

[asimov]

Nice work Stuart. My eyes like the first one best. There's some storm detail, not visible in the 2nd image.

[Saturn%5]

Very nice Stuart, I like the first one best :)

cheers

Graeme.

[U.K.Cowboy]

Thanks guys! You are dead right John the single stack avi is sharper. De-rotating images works well with poor to average data but in good conditions single avis seem to work best.

[asimov]

Agreed Stuart. I've not bothered about derotating as yet, in fact I only deleted Winjupos off my computer the other day without trying it lol. With Saturn in rugged seeing I simply capture for longer & forget the concept of any globe details like storms etc.

[U.K.Cowboy]

Deleting winjupos.......such a rebel John! ;) Look what you are missing out on, this is a North polar projection showing the hex storm quite nicely. This was combining 2 avis taken at 21.41 and 21.48 UT when seeing was at its best.

[asimov]

That's ok I'll just watch the experts play with their polar projection stuff hehe.

[U.K.Cowboy]

:lol:

[EricB]

Nice job Stuart. I like the first one best too.

By the way, how do a polar projection?

Cheers,

Eric

[U.K.Cowboy]

Thanks Eric. In winjup measure your image in "image measurements" then select "analysis" > "Map computation" and load up your measured image and then select the type of projection you want and click "compile image" and bingo!

[Quark]

Congrats Stuart, excellent result.

Let me start by saying I am a big fan of WinJUPOS derot. That said processing is an individual thing and there are many different paths that can arrive at a similar destination.

Blinking these two images, the level of noise in the single avi makes identification of discrete detail difficult. What is noise and what is real.

Looking at the derot image, clearly there is better signal to noise and the thing that caught my eye was the bright spot in the NEB that it resolves. This spot is pretty well in the middle of the NEB and a little toward the P limb.

I find that movement helps to identify detail. When you click on the image from your IIS thread it opens in a smaller window. If you drag that window, slowly to and fro across your screen the discrete detail is more easily identified.

In both your single avi and 20 min derot version this bright spot can be seen within the NEB. At first glance it is difficult to see but if you drag it to & fro across the screen it becomes quite apparent.

I have heard elsewhere this crock that detail is lost with WinJUPOS derot. Here you have an example of a very diffuse spot in the NEB that is visible in a single avi and also in a 20 min derot version. The thing I would point out regarding that fact, is you have retained a very diffuse feature in a much smoother image. Not smoother due to any smearing but smoother due to a greatly improved signal to noise ratio. Noise is not detail.

Regards
Trevor

[EricB]

Thanks Stuart. I will give it a try.

Eric

[U.K.Cowboy]

Thanks Trevor! I've been putting some images from that night over a 30 min period together for an animation and I think I can see that NEB detail. I need to figure out how to do a animated GIF as my usual youtube option is too rough for this subtle detail.

The Hex is rounder on the derot image (when I checked the polar projection) Trevor though maybe I should be splitting the 6 min avis into 3 mins and running those stacks through winjuop, maybe the 6 min stacks are causing some rotational blur do you think that could be possible?

[Quark]

I think consecutive short captures work very well. Some food for thought Stuart, check out this R channel I have attached. This is the R channel from 3 consecutive RGB data sets. Each channel was captured for 93 seconds and as I am in the habit of always starting the R channel on a half minute increment, that means there was a 1 minute gap between the first & second RGB as well as between the second & third RGB.

From the start of the first R to the end of the final B is 17 minutes. From the start of the first R to the end of the third R is 13.5 minutes. If you look closely at this R channel image the hex is very well defined as is the fine detail across the image, there is no smearing. This particular image was used as a reference image early in May by a professional research team imaging with Subaru and NASA's IRTF telescopes in Hawaii.

Of course using a colour camera will be different, however I think what I have demonstrated with this image would suggest that multiple shorter runs with the colour camera should be worth a try.

I cannot emphasize enough how critical the placement of the outline is in WinJUPOS that creates the IMS files that are then used to combine & derotate your individual data sets.

Good luck with it.
Regards
Trevor

[U.K.Cowboy]

Thanks Trevor I've got about 80 Gigs of data to wade through from this week of rare golden weather here in the UK so will experiment with winjup over the coming months until Jupiter emerges in the Autumn.

When I measure an image I raise the gamma over 2 to make sure I'm not cropping the outer edges with the outline.