imagej MER basics - Page 4

Previous 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 Next
Author Message

Posts: 3465

Reply: 61

PostPosted: April 21, 2012 1:11 PM 

Don, the Planetary Image Atlas returns 8 bit results for ALL the file formats allowed.

If you want to work with real original MER data then you must download the img files from the Analyst Notebook.

There is an imagej plugin ( PDS Reader ) that reads img files - including some of the 32 bit data. I use a version I modified to read all the img file formats that I routinely download - including the RAD ( Radiometrically Corrected ) files.

Furthermore, the planet day number must be 2700 or less for Opportunity - same limit as the images in the Analyst's Notebook.

Today is 2930 for Opportunity. Do you really want to wait 230 days to see the latest MER images? I thought not.


Posts: 3465

Reply: 62

PostPosted: April 21, 2012 2:04 PM 

sol 2699 ( Aug 28, 2011 ) comparison of JPG vs IMG file processing:

Can you tell which one was created using the 8 bit "raw" JPG data and which one was created using the 12 bit original IMG data?

OK, here's the deal: MOST of the MER images have already suffered the indignity of ICER compression before transmission to Earth. So not much gain there using the original data.

MOST of the images have a narrow brightness range so MOST of the images do not suffer many more slings and arrows when "stretched" to 8 bits ( less than a fraction of a percent of the image is "clipped"). So not much of a gain there.

So MOST of the time I do not bother to download the original data when it is available because MOST of the time there is little increase in dynamic range or acuity over what I can tease out of the 8 bit JPG images.

After all, the final products destined for the web MUST be converted to 8 bits! ( RGB is 8 bits each channel )

BTW, I keep all my intermediate processed images as 16 bit TIFF data on my computer.


Posts: 73

Reply: 63

PostPosted: April 21, 2012 4:23 PM 

Er... I cannot tell the difference. So why the extra bits if you dont need them. That being said, How often do we get RGB or L456 data to work with. Seems to me to use Hort's terms, it is mostly a dogs breakfast. We get L0, R1-R234, M1-whatever, 8 bit, 16 bit, and in the voice of Forrest Gump, some dont have any bits at all.

Give it up Don, The spotlight is on the Themis Man.....

Don Davis

Posts: 12

Reply: 64

PostPosted: April 21, 2012 7:52 PM 

Yes the IMG files are the best, I went through the NASAVIEW routine many a time. I don't mind 8 bits per channel, it works for digital camera photos of great subtlety.

The sundial pictures are traditionally the best bet with the jpegs because of the brightness range they give to the image, often resulting in better survival of the gray scales of the entire view. Images without the dial often become quite contrasty. I suggest comparing the same views from the PDS and the JPEG releases, case by case.
if your software is Mac compatible I might give it a try.
I have pointed out options for better quality material to process, and with that I move any further Mars discussion to the Cape York Mark II thread.


Posts: 73

Reply: 65

PostPosted: April 21, 2012 8:00 PM 

Dont rush off brother. Makes me think you are feeling the heat. This is how we move forward. The clash of the titans.

Yep i like it. aint like rocks is it? Where is the rock guys? They aint got trhe nads.


Posts: 73

Reply: 66

PostPosted: April 21, 2012 8:24 PM 

For those following along at home you should know the grandeur of the man in our discussion.

A link to show our respect,

NASA Mars Odyssey THEMIS Image: Mars in True Color (almost)

**** I fixed your link. The buggy comment parser strikes again.


Posts: 3465

Reply: 67

PostPosted: April 21, 2012 8:36 PM 

Don, if you click on the imagej link in the first comment of this topic - and then click on the download link you will find links to the various versions of imagej - including the Mac OS X version. The program is platform independent.

Yes, some images benefit greatly from the extra 4 bits of depth - mostly images with interesting details in both the shadows and direct sunlight. Such scenes also benefit from high dynamic range image processing.

The sol 11 image in reply 55 of this topic was HDR processed as there was interesting information across the entire 12 bits of depth. Even the 12 bits was not enough to prevent pixilation on the dark deck. ( If you look closely you can find the areas where this occurred. )

Anyway, hope you download imagej and give it a try. It is well worth the price ( free!! ) and well worth the investment of your time to learn how to use it.

If there is any interest beyond this introduction then I will post some useful macros for routine processing.


Posts: 73

Reply: 68

PostPosted: April 21, 2012 8:37 PM 

It would seem it just slips into error. The truth is that the Themis Mars images from Odyssey were given to Don Davis to colorize. It is mars art by Don Davis.

Good to have you stop by Don


Posts: 73

Reply: 69

PostPosted: April 21, 2012 8:45 PM 

It would appear he wants to run from Hort. We will see how it goes. Maybe we can have a long debate. We will see.



Posts: 3465

Reply: 70

PostPosted: April 21, 2012 8:50 PM 

Fred, why do you always "go there"? I am amused and annoyed simultaneously - just as Schrödinger's cat is both alive and dead at the same time.


Posts: 73

Reply: 71

PostPosted: April 21, 2012 8:51 PM 

Link in reply 66 seems to work now. Go figure.


Posts: 73

Reply: 72

PostPosted: April 21, 2012 8:56 PM 

Hort, just trying to stir debate. Hope Don dont run....


Posts: 3465

Reply: 73

PostPosted: April 21, 2012 9:06 PM 

There is no "debate" in imagej processing - just "doing".

Sounds like you need some imagej homework to keep you out of trouble...

How about downloading and installing the imagej Turboreg plugin and learn to register images for animations? ( You will laugh when you realize how simple it is. )

You might start by learning to use the built-in plugin "GIF Stack Writer" located in the Plugin / Utilities menu.


Posts: 73

Reply: 74

PostPosted: April 21, 2012 9:11 PM 

Er.... Yes professor....


Posts: 73

Reply: 75

PostPosted: April 21, 2012 9:39 PM 

In my defence. The semester is ending over the next three weeks. Much to do there. I will be free after May 15th or so.. Working on a Doctrine brother.....

AKA just a stiff that post here......


Posts: 3465

Reply: 76

PostPosted: April 22, 2012 10:43 AM 

Spirit sol 561 ( Aug 1, 2005 ) HD crop of Gusev L257 panorama:

with a link to the full panorama.

This image was done using the original data.

( You know I couldn't let Don's challenge go unanswered. )


Posts: 3465

Reply: 77

PostPosted: April 22, 2012 11:12 AM 

Speaking of the dead...

Spirit sol 2177 ( Feb 16, 2010 ) saturated false color of curiously shaped rocks in the soil of Troy in Gusev crater, Mars:

Several special processing steps were done in imagej: registration of the L2, L5, L7 images and a square root transform of the saturation. These techniques will be covered in later lessons.

I miss Spirit.


Posts: 1661

Reply: 78

PostPosted: April 22, 2012 5:52 PM 


I'm here-very busy this past week...but read your 47 including icer and jpeg sites with thanks. Lots of dynamic info in there that requires re-reading several times...but its the application that bothers me a bit.

Is that a 'stoned' screw shell in your 77 at 7 o'clock from center point? Sure damned well looks like it.

Macros time are a bit off at the moment. I may try the method this week.

I have made a few new colored merges today and will post tomorrow.

I trust others will stay clear of this lessons thread you have created for us and keep their comments for another shouldn't be a place to make comments unless its from those that are taking your lessons. Too much annoying bumph to read.



Posts: 3465

Reply: 79

PostPosted: April 22, 2012 7:49 PM 

sol 2524 ( Mar 1, 2011 ) montage of colorized MI of peculiar features on the rock "Ruiz Garcia" on the SE rim of Santa Maria crater:

The images are 4X the registered average of 4 separate original 12 bit data MI images.

The colorizing pancam image was created from L2...L7 filters of 12 bit data.

Imagej was used for all the processing.

This was an exercise on seeing how much better the MI images would look when created from original 12 bit data.

Compare this image with this one using JPG 8 bit images.


Posts: 1661

Reply: 80

PostPosted: April 22, 2012 10:21 PM 

WOW! horton>>>tremendous visual difference. My comp has 2-1G Nvidia Geforce GTX 550 Ti video cards in SLI and your 79 comes in clear and sharp.

I'll keep my eye open for 12 bit data when scanning the images for imagej input. Thanks.


Previous 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 Next

Join the conversation:

Very Happy Smile Sad Surprised
Shocked Confused Cool Laughing
Mad Razz Embarassed Crying or Very Sad
Evil or Very Mad Twisted Evil Rolling Eyes Wink
Powered by MTSmileys