Printable Version of Topic

Click here to view this topic in its original format

Unmanned Spaceflight.com _ Tech, General and Imagery _ Eagle and Endurance DEM's available!

Posted by: djellison May 17 2006, 09:30 AM

http://anserver1.eprsl.wustl.edu/anteam/mera/mer_osu.htm

Two things..

1. Phil - YAY - bundle adjusted references for all the sites in the top two txt files!!!

And 2. Um....what does one do with those Bin's. Damn I wish I could code...if anyone knows a means to get that lot into any sort of appropriate 3D format they win the UMSF 'special cookie' award smile.gif

Just for fun - I bunged the XY data of MERB into Excel. It's not 'square' i.e. it's not accurate - but I think I'll go for something accurate in Photoshop smile.gif

Also - Z axis against Sol for Spirit...I've seen that graph somewhere else...but making it ones self from data is always much more fun - you can clearly see Bonneville, Missoula, West Spur and the return to Larrys Lookout.

Doug

 

Posted by: helvick May 17 2006, 04:35 PM

QUOTE (djellison @ May 17 2006, 10:30 AM) *
And 2. Um....what does one do with those Bin's. Damn I wish I could code...if anyone knows a means to get that lot into any sort of appropriate 3D format they win the UMSF 'special cookie' award smile.gif

Well the .bin's are dead handy for converting directly to a graphical bump map, creating shaded relief maps and so on, I'm certainly going to have a bash at the latter later. Converting to a mesh requires some processing though and that's outside of my area of expertise at the moment.

Posted by: djellison May 17 2006, 05:16 PM

I've begun to make some progress - it's getting quite interesting smile.gif Some piccies of Endurance later perhaps biggrin.gif

Doug

Posted by: helvick May 17 2006, 07:52 PM

QUOTE (djellison @ May 17 2006, 06:16 PM) *
I've begun to make some progress - it's getting quite interesting smile.gif Some piccies of Endurance later perhaps biggrin.gif

There's some oddness in the labelling of the files - the Eagle bin file is labelled as 68Mb but is 29,160,000 bytes (2700x2700x4bytes per data point), the endurance bin is labelled as 28Mb but is actually 1312416 bytes (588x558x4).
Quick and dirty first pass at the Eagle bin file - converting to png. Apologies for the garish colour scheme but its a quick and dirty linear colour ramp just to see if the data all works.

I'm having some problems with the endurance bin - I'm not getting anything sensible out of it yet.

Posted by: helvick May 17 2006, 08:03 PM

QUOTE (helvick @ May 17 2006, 08:52 PM) *
I'm having some problems with the endurance bin - I'm not getting anything sensible out of it yet.

Just noticed that the eagle bin is big-endian (Motorola FP32) and the endurance one is little-endian (Intel FP32). Odd thing is that the eagle one works on my Intel machine and not the endurance one. Hmmh.

Posted by: helvick May 17 2006, 09:00 PM

Something definitely odd - the file labelled Motorola FP32 (Eagle) is actually littlendian and the Intel FP32 (Enfurance) is bigendian. Maybe it was just a test. Anyway as "grayscale" images - got to work out how to output the data with better depth resolution (these are true color pngs just using gray values so there's really only 8 bits of depth) which is about 1cm for Eagle and 10cm for Endurance. The data for Endurance isn't quite square but the error is only about 0.07% so it amounts to less than half a pixel over the full image so it's not noticable.
Endurance:


Eagle:

Posted by: djellison May 17 2006, 10:46 PM

Obviously there are artifacts in the Endurance DEM - so I've used a mesh density half that of the actual DEM to help cull some of that - after all there is a disclaimer on the page saying that once they get a chance to sit down and catch their breath they'll get these tidied up etc. Who knows...we might end up with one of the Southern Basin, or Victoria Crater smile.gif

I think probably the most appropriate 'res' for something like this lies between the two densities...something like 1 million points is essentially 'plenty' smile.gif

Doug

 

Posted by: helvick May 17 2006, 11:01 PM

QUOTE (djellison @ May 17 2006, 11:46 PM) *
I think probably the most appropriate 'res' for something like this lies between the two densities...something like 1 million points is essentially 'plenty' smile.gif

I've run into a wall with trying to push the data out in a decent 16bit grayscale format using the quick and dirty approach. I'll try again tomorrow with something a bit more capable than Perl::GD. I'm still a bit confused about the big endian \ little endian thing though doesn't make much sense.

Posted by: CosmicRocker May 18 2006, 04:19 AM

Nice work with that last one, Doug. I need to refresh my skills with this grid plotting software, as I didn't create anything as pretty as your view. I'm working with the ASCII files, since they are easily accepted by my software. I can see that there is quite a bit of detail in the data; now to learn to bring it out as you have. I tried a shaded relief map of Endurance, but it was difficult to make it look well, and it appears to have a strange network of artifacts. I wonder if that is some of what they hope to eventually clean up.

I found it curious that the ASCII file for Eagle was huge...60+ MB, while that for Endurance was relatively small.


 

Posted by: djellison May 18 2006, 07:00 AM

I used the Ascii files with Surfer, but then 3dem with the bin files to generate a Terragen mesh from which I made a Tif which I used as a displacement map smile.gif

There's what, 250k points on the End one, 7.29 million in the Eag one - so it's 30 times more data than the End one...not suprising it's a bit big.

I wondering if perhaps, just perhaps this little easter egg of data is in response to the email I sent to Prof Li a few months ago asking if he could do elev maps for us nutters smile.gif

Attached - the dem at the best res I've got it into 3ds, and the res I actually use it at smile.gif
Doug

 

Posted by: djellison May 18 2006, 10:21 AM

6.6 Megish...
http://www.unmannedspaceflight.com/doug_images/beta_end_dem_phil.wmv

Not happy with the compression though - I'll try a H264 Mov at 720p and half 720p later smile.gif

Doug

Posted by: Pertinax May 18 2006, 01:11 PM

Beautifull work -- particularly nice with the overlay.


-- Pertinax

Posted by: Phil Stooke May 18 2006, 01:22 PM

Very nice maps. The artifacts in the Endurance map must be caused by the interpolation routine they use to grid the data.

Phil

Posted by: aldo12xu May 18 2006, 04:03 PM

Great piece of work, Doug. In the video, it really becomes clear that the rocks from Escher to Wopmay are part of a slump feature. And it's possible to follow the contact between the Upper and Middle Units of the Burns Formation. Really cool!

Posted by: algorimancer Jun 24 2006, 01:14 AM

QUOTE (djellison @ May 18 2006, 05:21 AM) *
6.6 Megish...
http://www.unmannedspaceflight.com/doug_images/beta_end_dem_phil.wmv

Not happy with the compression though - I'll try a H264 Mov at 720p and half 720p later smile.gif

Doug

Very nice. Funny, the bowl of the crater has much shallower slopes than I'd anticipated... my perception had been that the rim was pretty steep. Good job of eliminating the artifacts.

Powered by Invision Power Board (http://www.invisionboard.com)
© Invision Power Services (http://www.invisionpower.com)