Make your own Jupiter from my data

iceman · #1 (**permalink**) 07-June-2007, 10:17 PM

Hi all

In the continuing trend of collaborative processing tips and techniques, open sharing of information and ideas, I've uploaded the raw stacked images from my excellent Jupiter session on the 26th May.

The files are "raw stacked". That means, I've ran them through registax to align and stack the raw frames, but have not done any wavelet or other processing on them. They have had mild noise reduction applied in photoshop (part of my usual processing routine).

Each file is a 1.2meg TIF file. Right-click on the link and click "save target as" to save the file.
- Red Channel
- Green Channel
- Blue Channel

As a suggestion for the less initiated, open each channel one at a time in Registax, and it will take you straight to the wavelets page where you can start the processing.

Use AstraImage, Photoshop or similar tools to combine the image back into an RGB image.

An (un-inforced) requirement is that you post your processing steps when you post your version of the image. This will help everyone (myself included) learn from your method of processing and be able to compare the results.

Have fun, and I look forward to seeing the results!

paul f. campbell · #2 (**permalink**) 07-June-2007, 11:24 PM

Hi Mike.

Thanks, this sounds like fun, I will give it a try. Thanks again Paul

Occam · #3 (**permalink**) 07-June-2007, 11:39 PM

Nice - thanks. Here it is with a minor tweak to boost the colour slightly
I used a fairly basic technique in Paintshop Pro 7
1- opened the three files
2- converted them to greyscale (necessary, since the TIF's are technically still 24bit/16 million colours)
3- Combined the RGB channels
4- Ran a simple fade correction at level 5

iceman · #4 (**permalink**) 08-June-2007, 12:41 AM

Well done Occam.

You might try using Curves to increase the contrast, and try some sharpening filters as well. There's a lot of detail hidden in the data, as in it's downloaded state, it's completely raw.

andyschlei · #5 (**permalink**) 08-June-2007, 01:25 AM

Quote:

Originally Posted by iceman

The files are "raw stacked". That means, I've ran them through registax to align and stack the raw frames, but have not done any wavelet or other processing on them. They have had mild noise reduction applied in photoshop (part of my usual processing routine).

Are the images aligned?

iceman · #6 (**permalink**) 08-June-2007, 01:35 AM

Maybe, maybe not.

What I mean by that, is that this is how they came out after my initial pre-processing, aligning and stacking.

My procedure from here then includes wavelet processing in registax, deconvolution in AstraImage, recombine into RGB in AstraImage, and then further post-processing in Photoshop.

So when recombining into an RGB image, that's where I check the RGB alignment. It's likely that 1 or 2 channels may need to be moved by a pixel or 2.

It's worth noting that because of the long focal length, and the long capture time (40s each channel), some features have rotated between the Red channel and the Blue channel.

So when aligning the channels, if you align on the "edge/limb" of Jupiter, the features in the middle won't be quite aligned. I always align on the features themselves, which sometimes leaves a blue rim on the right hand side but it's more important (to me) that the features are aligned and accurate.

If you use the "channels" tab in Photoshop (once it's an RGB image), you can alternatively click on each of the Red, Green and Blue channel layers and look for any movement in a particular feature (eg: a white spot, Red Jr etc), and use the Move Tool to re-align each channel.

Hope that helps.

aerovoid · #7 (**permalink**) 08-June-2007, 03:31 AM

My first attempt. I never really understood what the wavelets did, till now. Since I don't have a telescope I don't really get to use registax much at all. But I adjusted them manually to see what they did and I did that for each image. After that I opened all 3 in photoshop. I pasted them all into one image, each on there own layer. And then right clicking on each layer and going into "Blending options". I leave the appropriate RGB box checked (eg, leaving just the R box checked for the red image). I then merged all the layers and then went into "levels" and adjusted them manually to what I felt looked good. I tried using some "Neatimage" to get rid of some of the noise, but it removed too much detail, so it wasn't worth it.

I'll probably try it again from the beginning again and see if I can do better.

andyschlei · #8 (**permalink**) 08-June-2007, 04:45 AM

Mike,

This is a great thing you are doing, giving us all a chance to work with your fine data.

Since you suggested Registax and Photoshop, I decided to be contrarian an use PixInsight, a currently free image processing tool. It has a pretty steep learning curve, but it provides great control over many image enhancing tools.

So, complying with your direction to share processing steps, here is what I did.

I ignored the direction to align each color channel. I know this is being lazy, but there was no easy way to do it. In the final image, I don't see differences in alignment for each channel.
I combined each color image into an RGB image, factoring each color evenly.
I extracted a luminance channel using "Extract Channel" in PixInsight. I used this to get to the best wavelets for the image, based on the tutorial on the PixInsight website.
Finding the right parameters for the Atrous wavelet tranform, I was still not satisfied. I created a mask using a severe wavelets transform of the planet. Parameters were: Levels 8 & 9 (scale 128 & 256) bias 1.0 and 0.1 respectively. All other layers disabled. I then stretched the resulting with curves, eliminating the low end, to create the final mask.
Using the mask, I applied wavelets to the entire image. The parameters were, by level: 1: off, 2: +12, 3: 0, 4:+0.8, 5:+0.6, 6: 0, 7: 0, 8:+0.3, Dynamic Range Adjust: High: 0.4
Still using the mask but using it inverted, I darkened the limb of the planet. This improved the "flattened" look created by the sharpening process
I applied an overall curves (no mask) that increased contrast and boosted saturation. The ability to manipulate hue and saturation in the curve dialog is a great strength of PixInsight.
I used the GreyCStoration noise reduction algorithm to smooth out the noise in the image. I increased the scale of the noise to 1.2 pixels, all other parameters were default.
I saved the file as a FIT (PixInsight works in 32 bit floating point, so saving in FIT saves all the data) and a TIFF for Photoshop (16 bit integer).
I saved the file as a PNG from Photoshop. No manipulation with that tool. I tried an high-pass filter, but it didn't really help.

I'd be happy to share all intermediate files or PixInsight processing details.

Thanks Mike for a great project!

--Andy

aerovoid · #9 (**permalink**) 08-June-2007, 05:48 AM

Ok I made another one already. I did the same things that I did before, but adjusted them a bit differently. Then I also did a Hue/Saturation adjustment. I only did a subtle saturation on the red and yellow and then I played with the curves to add more contrast. I think it looks much better than the previous one I did.

stargazer_7000 · #10 (**permalink**) 08-June-2007, 07:17 AM

hi Mike,

thanks for the opportunity to process such wonderful data!
I`ll give that a go this weekend!
great idea!!!
(-:

JAICOA · #11 (**permalink**) 09-June-2007, 09:04 PM

Hi Mike
I had a kick on this, Great Idea for everyone to share
First i did not used Registax i went straight to PhotoShop cs2 and performed the following.
1. went to Filters selected UnSharp Mask moved the sliders Accordingly
a. Amount 35%, b. Radius 4.5% c. Threshold 0 - on all

2. Filters Selected Gausian Blur .4

3. Image/Adjustment selected Curves gave the "S" shape until my liking on the edges and center.

4. Image/Adjustment selected Levels and rise it to 15%

5. Then selected Filters and went to SmartSharpen move the sliders to
a. Amount 30% b. Radius 4.6% c. and Selected Lens Blur

6. Filters Selected Gausian Blur .6

7. then went to Image/Adjustment and Selected Saturation increase to 26%

8. and also Selected Color Balance and went to ALL three (Dark,MID,HIGH) and selected ONLY the YELLOW/BLUE Slider to Yellow 5%

9. Adjusted Brightness Slightly and Contrast

10. Repeated Step #1

11. Repeated Step #5

I Hope i got Everything that i could remmenber.

Mike seeing the raw data you have some beautiful skies and a height position for this Planet. I have been trying but too low and too much turbulance.
Took some sessions last night will post when processed. And thanks again for letting me and everyone else Participate on this Great Idea.

Bokmakierie · #12 (**permalink**) 10-June-2007, 08:03 AM

Excellent Efrain!! I realize now how little I know when it comes to image processing. Will have to take an intensive training course in Photoshop!

Phil

Galactic2000 · #13 (**permalink**) 12-June-2007, 03:18 AM

Hi Mike,

Here is my crack at processing your excellent Jupiter images.....

First Note....
this was a little bit of a challenge as the blue channel -Jupiter has rotated a significant amount compared to the Red and Green Channels. This accounts for the difficult color balancing of this image.

I first started out in Maxim DL,
using DDP Kernal Low Pass filter
or you can also use
Mild FFT Low pass 5% cutoff, with 50% weight mask

next in Maxim DL
Local adaptive filter, Radius 9 and Contrast 2
Next
Convert all three to Grey scale images,
align in MaximDL via align command, or align during Combining,
Combine RGB 1:1:1 ratio, Auto correlation, for aligning.
Realign Color Planes to tweak the blue channel into its proper place.

Save as RGB, .tif

Go to Adobe PS, adjust color balance, and levels, save as Jup_RGB1

Go to Registax 4 run thru wavelet filter, use Linear, (Not Gaussian) tweak middle wavelet slider slightly.

Adjust brightness and contrast and
save as Jup_RGB2,
Go to a Adobe and choose filter, despeckle .1, save as Jup_RGB2

Luminance Layer the RGB1 and RGB2 in ADOBE with RGB2 as background and RGB1 as Layer. Tweak % on both to get a happy medium in sharpness and softness.

I tried Deconvolusion and was not happy with the result, so I stuck with the above process.

I tried many versions, but this is what I finally decided on.

Mike,
Thanks for the opportunity to play with such excellent Data...I rarely get seeing this good in Ohio, so thanks for the good time.

You truely are the "Planet Master"!!!!

snabald · #14 (**permalink**) 12-June-2007, 04:47 AM

Photoshop 5.0

1 - converted each image to 8 bit grayscale
2 - opened new file same dimensions as 3 files
3 - pasted each file into respective channel.
4 - set overall contrast to +20
5 - manually set input levels to 24, .067, 221 on RGB channel.
6 - ran sharpen filter at default level.
7- saved as JPG quality level 9.

Galactic2000 · #15 (**permalink**) 12-June-2007, 05:35 AM

Something went wrong in my previous image alignment in Maxim, so now I got it right it looks beautiful now!

I started out in Maxim DL,

Convert all three to Grey scale images,
align in MaximDL via align command, or align during Combining,
Combine RGB 1:1:1 ratio, Planetary, for aligning.
next in Maxim DL save the RGB,
use
Local adaptive filter, Radius 9 and Contrast 2
Next
using DDP FFT Low Pass filter
or you can also use
Mild FFT Low pass 5% cutoff, with 60% weight mask
Save as RGB, .tif

Go to Adobe PS, adjust color balance, and levels, and do an unsharp mask and save.

Bokmakierie · #16 (**permalink**) 12-June-2007, 09:40 AM

John, this is absolutely magnificent! Congratulations.

Phil

iceman · #17 (**permalink**) 12-June-2007, 01:38 PM

Hi guys

Apologies for the late reply into this thread - i've been away all weekend and wanted to see the images on a proper LCD instead of the laptop screen.

Thanks for all your efforts and explanations - it's great to see how people treat the same data with their favourite programs.

I'd like to comment on a few of the examples - please don't take my comments as criticism. I'm just hoping to share my opinion in order to hopefully help you become a better image processor.

Andy - your version appears to be a little blunt on the noise reduction and sharpening, plus it's lacking colour and has ended up as browns and whites with not much in-between.

Erfain, I really like your version - lots of detail visible though on this screen it appears a little bit dull and the edges are a bit sharp. Check the levels in PS2 and perhaps try lifting the white point a bit, and maybe a feather around the edges (3px or so) with a guassian blur might help soften the hard edge. Otherwise very nice, I like it a lot - and I like how you used CS2 only - I can learn a lot from your processing.

John - yours is very nice also, quite similar to how I get mine to look though I do capture with a high gamma and high gain so the image can look washed out if it's not reduced in post-processed. Your version appears (to me) to still be a little washed out and could probably do with some curves adjustment but I do love the colour and detail.

By way of comparison, here is my version of processing on the same data, done back at the time of capture.

Of course, processing is a very subjective thing and what looks good to one person might look oversharpened, or the wrong colour, or whatever, to someone else.

I hope this project has helped anyway, and I'm happy to do the same again in future when I've got good data to share.

iceman · #18 (**permalink**) 12-June-2007, 01:47 PM

Efrain - as an example to what I meant, I took your posted image and adjusted in Photoshop CS2 in the following way:

1. New levels layer, and adjusted the white point individually of each channel, sliding it back to the left until it reached the data, and then to taste for the colour.

2. For each channel, selected the outside of the planet, inverse and feather about 3 pixels, guassian blur on the edges.

3. I also found that the R,G,B channels were not properly aligned with respect to the features so I shifted the channels to align the features.

That's it.