Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Jupiter processing challenge.


Space Cowboy

Recommended Posts

  • Replies 70
  • Created
  • Last Reply

Okay, I had to bite as I'm a sucker for image processing. Here's my go. Wavelets in Registax and an Unsharp mask added in Photoshop. Whilst in Photoshop I also thought the negative looked quite good!

post-31280-0-09210400-1389903231.jpg    post-31280-0-89931300-1389903233.jpg

Rob

Link to comment
Share on other sites

Some really interesting versions here. Can really tell a lot from this. Can tell peoples style of processing big time. I particularly like Riklaunims fine appearance. Though I think sizes should be kept the same or similar to the original.

And Petes image is the complete maverick of the bunch. Showing detail on the poles better than any body elses. But the balance seems quite dark and rather different to what I am used too.

I thought that was the high frame rates that Pete has been using. Clearly I am wrong about that. And infact its hes processing. I hope at some point Pete could share the processing. I am particularly interested in the dark features that he has got out. Just wondering if similar can be done. But with a more neutral balance ? Never to old to learn new things. And we are all learning from this I am sure. Even Pete I would have thought.

Infact I think I can see pros and cons in most of the images. Some things that look right. some things that clearly look wrong, and often in the same image. Including my own.

To be honest I think I have the most to learn from Pete. I wish I could get that dark pole detail. But with a balance closer to my own. As said Pros and cons. But Pete, I think is showing detail in places that are missed by most on here. Including my own.

Interesting Stuff.

Mr Davenport that started this off with hes comments. Clearly has a lot to learn from many of us on here. Rather too contrasty dark and noisy for may tastes. But Then many experienced users on here, also have problems that stick out to me. I wont go into all of them. But there clearly there. And that includes my own I guess.

Good Job Everyone. I bet your surprised by this Stuart. I certainly am. It really shows what we are all doing in the stark light of day.

I hope Stuart Puts up hes version. For the microscope. Someone said it felt like a exam. I thought that too. :grin:

post-2700-0-13118300-1389952822.png

Link to comment
Share on other sites

Thanks Neil,

It has been very interesting indeed and I appreciate the feedback, it is difficult teasing out detail without getting to much contrast and vibrance!

I take it Pete has won :rolleyes:  - Well done sir!

I would like to vote, Chris from Cambridge for second place! 

I must say I think in the nervousness of the moment and hurrying through the process I have pushed everything a bit further than it needed to go.

This is a typical Noob thing and good learning for me.

Second learning is noise control, how does Neil get the detail with out the additional noise?

My attempt was a pure Registax and Photoshop attempt (cheated with the brightness of my moon :)

I call upon both Neil and Chris to, at least, share the software and features used? (Please)

Thanks to everybody, exciting stuff that has inspired me to go trawl back through my data and to reprocess!

(At least something fun do until the foul weather passes!)

PS: I like the fact that the moon is a doughnut and so is the shadow, take it this must be a collimation issue?

Link to comment
Share on other sites

Busy day here so just a quick comment. Yep Pete has popped some extra detail for sure I reckon he's utilized the red channel more than the others.

Hold your horses Chris I'm judging this one! The shadow is from Ganymede which is out of shot , my collimation indeed! :shocked:

Anyone can have another go if they are not happy with their first attempt. :grin:

Link to comment
Share on other sites

I have say that I agree with Neil and am really impressed with the detail Pete has extracted, especially at the poles.

From memory my processing was something along the lines of:

1. RegiStax 6 - RGB align.

2. RegiStax 6 - Wavelets (just one of my standard saved wavelet sets, not tailored to the image at all).

3. Image Analyzer - A little noise reduction

4. PixInsight - Tweaked curves, including boosting the colour saturation.

5. Photoshop - A little unsharp mask on the moon - is this cheating?

6. Photoshop - Auto Colour - possibly a mistake now that I look back at the intermediate images!

Cheers,

Chris

Link to comment
Share on other sites

Stuart,

can't decide if this is a sly trick to offload the processing, an elite competence test to benchmark skills or a competition you have yet to announce the grand prize for ;)

All tongue in cheek and no suggestion of nefarious intent!

However it's an extremely interesting, informative and fun thread - great idea and I shall have my own stab to make everyone else look good.

Link to comment
Share on other sites

Has Stuart invented a new sport here??!! Though I think there should be some rules as we don't know how long people took and as Neil says Riks images look nice but at full scale look completely different. It's all been good fun though and good to see people posting up their images.

I found this very interesting doing a process of someone else's data as I think you get used to your own data and how to process it. With this I did RGB align for the first time in years (usually do mono) as did Chris and I'm sure all the other experienced imagers, but when I applied wavelets and then moved onto further processing, I realised the wavelets weren't quite right on this data. IA was also different with OSC data. Maybe I should get my DBK out more often to try some different processing. Bought back fond memories of processing my DBK data though.

What chance of this becoming an Olympic sport in 2016??!!!!

Link to comment
Share on other sites

Interesting point Neil makes about the darkening of Pete's image.  I often darken mine and in retrospect whilst I do it because the final image appeals to me more, I've never actually thought about why.  Is it perhaps because the contrast increases (and perhaps in Pete's version this is why more detail shows at the poles)?

James

Link to comment
Share on other sites

Sorry can't edit yet, let me rephrase the above as it is not my choice!

I propose Neil is in the lead and I propose that that Chris is coming second.

Just realised that both James, Stuart and some others have not posted images yet.

But will they rise to the challenge?

You know. I suppose when someone's process is balanced right across the board, Meaning more detail. less noisy. Correct colour, good levels. ect ect  Then perhaps we could say there might be a clear winner. But to my eyes things are not quite so clear cut. What appeals to one. May not appeal to another. Are you the winner if you have the most detail ? What about the most detail but only at specific shades ? does the other processes that might show detail in a different way. Not get merit ?

 If only detail was the goal. Colour wouldn't matter. Noise wouldn't matter. Smearing of very fine detail wouldn't matter. Levels wouldn't matter on and on. As such it feels short sighted to me. to talk in terms of winners and losers. I am more interested in the strengths and weaknesses that all images are showing on here. Some may have less weaknesses for sure. But I don't believe there is a jack of all trades possible. Some trade off will have to happen somewhere. Even in the the top images processed here. If I see a image that has everything. ill change my mind about a winner. But so far I haven't seen it. Regardless that there is excellent processes done.

But its nice for people to say good things about your processing. So cheers for that.

As for detail and noise, I do use frequency domain on image analyser. I also use spline filter on registax. very careful sharpening only using what I think is needed. No more no less. And good wavelet control. Think I used Linear second wave only registax 5, no guassian. wavelet around midway somewhere. ( I didn't log it ) but I resize the image to 150% on registax  and look at the wavelet noise level like that when I apply wavelets. It helps the eye see even tiny amounts of noise appearing as the wavelet is pushed up.

Link to comment
Share on other sites

Do we need a winner?  I think it's very interesting to compare the images.  Some bring out more detail in one place, others in another, yet others have a slightly different colour balance depending on what floats their boat.

It's a bit like judging what's the most "natural"-looking.  Has anyone been there to check out what it really looks like? :D

James

Link to comment
Share on other sites

Busy day here so just a quick comment. Yep Pete has popped some extra detail for sure I reckon he's utilized the red channel more than the others.

Hold your horses Chris I'm judging this one! The shadow is from Ganymede which is out of shot , my collimation indeed! :shocked:

Anyone can have another go if they are not happy with their first attempt. :grin:

You may be right about the red channel. When I did single shot colour. I started using a extra red added in. Didn't try that here. Just a straight RGB . If that's how Pete did it ?  ( feels like more than that to me ) The darkness is a clue ? Some kind of levesl histogram trick me thinks. Though another red could be the culprit too ? The colour balance isn't quite right. Which often happens with extra channels as we all know. Hoping Pete tells all. Got me thinking all about high contrast Reds Now.

Link to comment
Share on other sites

Btw Think I should spill the beans about the moon. I processed it separately and pasted it on. I know others have too. I resized one image and saw the square box. ( tip blacken a touch afterwards )

 It couldn't have those healthy levels with out. Also did a RGB align on the shadow only.  And added it back it using layers on Gimp. What surprised me is how good the shadow looks on a lot of the shots. Its easy doing a colour noise reduction on it. and having little colour in the Jupiter image. But as soon as you start having some nice colour in the image. without doing a RGB align the moon shadow goes technicolour. Its quite a challenging image Stuart. Good choice. I am surprised how good the shadow looks on a lot of the shots.

Link to comment
Share on other sites

Ok - not wanting to detract from the more carefully processed - here's my stab with R6 and Gimp 2.8.10

Drizzle15_4-46-35-78-Part0000_pipp_g4_b3_ap2r6.tif

I will revisit later as I think pushed this too hard to get the polar detail, at the cost of increased noise and equatorial brightening.    I would however be very happy if I could capture some of my own data of this quality ;)

Ok no in line tiff - here's the same as a png:

post-26731-0-53087300-1389962038.png

Link to comment
Share on other sites

Do we need a winner?  I think it's very interesting to compare the images.  Some bring out more detail in one place, others in another, yet others have a slightly different colour balance depending on what floats their boat.

It's a bit like judging what's the most "natural"-looking.  Has anyone been there to check out what it really looks like? :D

James

Exactly James. Getting everything at premium, is almost impossible. there has to be weaknesses somewhere. One might have more detail. one might have better colour balance and levels. one might show more easily seen dark contrast, another might show slightly finer detail. with a little less smear.

And as you say, its all subjective to how it should really look anyway. The most one can say is , I prefer that one, or this one. 

Link to comment
Share on other sites

This is my effort.  I'm finding it really hard to keep the noise in the blue under control.  I think splitting the colours and processing the blue to get rid of some of it first is probably the way to go.

post-10871-0-40217400-1389963318.png

James

Link to comment
Share on other sites

Good points on what might be the "winner" here guys. As you say, even the "best" images have their issues. I think there are however some basic techniques being used such as RGB align in Regi, noise control etc that hopefully the less experienced imagers can pick up on and apply to their own images. I think everyone is learning something from all these images as I don't think you ever stop learning in this game.

Link to comment
Share on other sites

This has been a great thread for learning about planetary imaging. When I get home from work tonight I'll have another pop trying to apply some of the techniques mentioned on here. Some fantastic efforts all round.

Rob

Link to comment
Share on other sites

You know. I suppose when someone's process is balanced right across the board, Meaning more detail. less noisy. Correct colour, good levels. ect ect  Then perhaps we could say there might be a clear winner. But to my eyes things are not quite so clear cut. What appeals to one. May not appeal to another. Are you the winner if you have the most detail ? What about the most detail but only at specific shades ? does the other processes that might show detail in a different way. Not get merit ?

 If only detail was the goal. Colour wouldn't matter. Noise wouldn't matter. Smearing of very fine detail wouldn't matter. Levels wouldn't matter on and on. As such it feels short sighted to me. to talk in terms of winners and losers. I am more interested in the strengths and weaknesses that all images are showing on here. Some may have less weaknesses for sure. But I don't believe there is a jack of all trades possible. Some trade off will have to happen somewhere. Even in the the top images processed here. If I see a image that has everything. ill change my mind about a winner. But so far I haven't seen it. Regardless that there is excellent processes done.

But its nice for people to say good things about your processing. So cheers for that.

As for detail and noise, I do use frequency domain on image analyser. I also use spline filter on registax. very careful sharpening only using what I think is needed. No more no less. And good wavelet control. Think I used Linear second wave only registax 5, no guassian. wavelet around midway somewhere. ( I didn't log it ) but I resize the image to 150% on registax  and look at the wavelet noise level like that when I apply wavelets. It helps the eye see even tiny amounts of noise appearing as the wavelet is pushed up.

Neil, 

Point taken, I am just edging people on trying to get more people involved.

(Imagine how hard it must be to judge best Astro picture of the year when you have 1000s of entries of many subject matter.)

I always get my wife to choose the best picture from a stack of options post a good evenings capture and process.

This is because I become to subjective and obsessive about certain aspects instead of balancing the overall look and feel.

Don't think I or many others on here really care who wins, because we have already all won by being involved and learning new techniques.

Link to comment
Share on other sites

It is interesting seeing all these results from the same data. I agree with you Neil about my image being a bit on the dark side and to be honest it's due to me being a bit cautious of other monitors being brighter. I remenber that issue I had with a Saturn from last year where you & Freddie were picking something up in the background that I just coudn't see on my monitor. From then on I probably hold back a bit to ensure that things don't look too bright on other screens.

I have made a few minor adjustments so hopefully this will look a littIe better from that point of view. 

Pete.

post-22012-0-21814700-1389967319.png

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.