Jupiter 12th Feb from tassie - final

davidpretorius · #1 12-02-2006, 10:33 PM

Hi, i find that i get cleaner / sharper final images if i go less agressively on the wavelets in registax and then a mixture of LR deconvolution (4x2) and unsharp mask (medium 2.5) and a 3x3 median. The beauty of image scale, seeing etc is that the detail is there to refine, not like mars for me where i had no idea what was there until i heavily prcoessed.

These last couple have been extremely pleasing. From the minute i watch the image on the laptop during capture i know what detail is there and all i do is refine it.

RB (Andrew) · #2 13-02-2006, 12:51 AM

Excellent David.
Detail and image scale is tops.

Looks fantastic.

bird (Anthony Wesley) · #3 13-02-2006, 03:49 AM

That is a really nice final image DP. I agree with your assessment on wavelets + deconvolution - it comes down to information preservation, gentle on the wavelets preserves more data in the image for the deconvolution to use.

I've found myself using LR deconvolution on red and green channels and VC deonvolution on the blue channel, mostly because the blue channel is a lot blurrier than the others and needs something more violent than the LR to sort it out.

regards, Bird

iceman (Mike) · #4 13-02-2006, 06:06 AM

Looks good DP, and as Bird said, that's what I find best as well, especially with the separate RGB processing.

The Raw mode colouring is certainly different than the colours i'm getting, which are much more coppery/red. I don't know what's "right", I guess personal preference plays a large part.

Look forward to your next night of good seeing.. Your results are most excellent.. tracking + good seeing = greatness.

davidpretorius · #5 13-02-2006, 08:17 AM

thanks bird & ice,

i have often had a look at VC, but will have a more serious application of it.

ving (David) · #6 13-02-2006, 02:46 PM

great work again dave

we have some really talented planetary imagers her, and you are one of them

Robert_T · #7 13-02-2006, 02:55 PM

Hi Dp, love your work.

What you describe re process is pretty close to what I've come to as well. Looks like a minor red spot in this one too?

cheers,

davidpretorius · #8 13-02-2006, 04:15 PM

thanks ving, that means a lot.

rob, yes i have noticed that mini grs in this one. it is apparent in the 5fps compare ones as well

asimov (John) · #9 13-02-2006, 09:58 PM

Thats a very nice image Dave!

robin · #10 14-02-2006, 09:58 AM

Very good David.Impressive stuff.Great image scale.Been away from the forum quite a bit lately so Im not entirely up to speed with performing deconvolution in Astra Image...are there any threads on the forum that give you a basic,step by step guide?

davidpretorius · #11 14-02-2006, 10:46 AM

Quote:

Originally Posted by robin

Very good David.Impressive stuff.Great image scale.Been away from the forum quite a bit lately so Im not entirely up to speed with performing deconvolution in Astra Image...are there any threads on the forum that give you a basic,step by step guide?

have pm you, but will post my reply here as well
__________________________
http://precons.com/iceinspace/Articl...otracking.aspx

have a read though towards the end of the article.

The article deals with a final image from regisax, which in astral image you then split in colour panes. and then work on each colour frame and then recombine OR// if your image is really good, then split the original movie into three separate movies r,g & b with a program aviraw and then registax each movie separately. They vary a heap!!!

Then bring each final image into astral image, convert each to greyscale and then proceed from there.

I find that if you apply wavelets to a "backed off" stage ie the detail is starting to show, but dont go all the way. deconvolution applied to this image produces a crisper and cleaner that a heavily waveletted process.

We (Bird, Rumples & I) will be going through this at snake valley on a projector screen i reckon with some great data of jupiter, so it would be great if you can still make it.

bird (Anthony Wesley) · #12 14-02-2006, 06:29 PM

Quote:

Originally Posted by davidpretorius

We (Bird, Rumples & I) will be going through this at snake valley on a projector screen i reckon with some great data of jupiter, so it would be great if you can still make it.

I was thinking of bringing my Jupiter data on a dvd and giving copies to anyone that wants to take it away for practise. Are you likely to have a dvd writer there DP, or anyone else? I've only got a CD writer in my laptop.

I wanted people to see the raw data so they realise how awful it looks - extremely short exposures give frames with lots and lots of noise, but then this noise magically vanishes when you stack a few hundred of them. I think a lot of people are trying to capture raw frames that look "good", but this is a mistake (imho), and the long exposures needed to do this are introducing blurring that can't be removed.

On jupiter I'm using exposures around 30ms, and through colour filters this makes the planet very dim.

Bird

davidpretorius · #13 14-02-2006, 06:42 PM

dvd r/w,

great idea. I hope to have a little recipe book done by then. It is just a matter of getting it all together

Sausageman (Mike Boggan) · #14 17-02-2006, 04:30 PM

That is one superb shot David.
Congratulations.
when I get my 20DA I hope to get some shots like that. With some help from Striker and 1 ponders of course.

Mike