I'm annoyed of the difficulties I had to identify far lanscape features on Spirit panoramas.
So I try to compute a mesh (I love meshes!) from MGS MOLA datas http://pds-geosciences.wustl.edu/missions/mgs/megdr.html , to represent the rim of Gusev as seen by Spirit. I use 128 pixel/degrees data, cropped gusev area, sphericise the heights according to Mars diameter.
Here is a first image computed for SSW octant horizon, compared to a JPL Spirit image (ma-adim_L7_sol161_p2375-A181R1.jpg). I'm amazed by the quality of far landcape rendering from MOLA data, though resolution is only~ 460m/pixel. Alas, all the small hills less than~20km far from the Rover are not represented (ie the flat foreground, but notice the flat big southwest mesas are present in front of the crater rim).
Anyway, I'm happy and hope to render and post a complete 360° synthetic Spirit horizon in the next days.
On the picture: the mesh used is figured at left, colors=altitude; red cross=Spirit location (estimate), other dots: remarkable landmarks also on lower right image.
Hope you will enjoy!

This looks great erwan, it is definetly a view I had hoped to see synthetically, so we can get a little perspective on what we're seeing in the stunning long-range images Spirit has been, and will be, taking. I'm having a little trouble putting your overhead map into perspective though, are the 3 dark blue marks the mesas north of the opening to to Ma'adim Vallis? And the green and light blue are the actual crater wall?

I'm surely enjoying! Nice work. Looking forward to see more.

Interesting work erwan but i would have enjoyed it more if you had posted that image in a smaller format, 2100x1045 pixels is too large for most desktops - around 800x600 is much better for forum posts or attach a link.

Thanks for the comments.

cIclops: You're right, page is far more easily readable now!

Here is a link to the original 2100x1200 image



Slinted: Yes for all your questions; now marks are numbered on right, as on bottom left.

Please notice the mesh reproduce only a north-south central portion of Gusev topography. Ma'adim mouth is just a little bit outside the mesh on SSE. I have marked with green the East and West bank of another valley mouth, located west of Maa'dim.

On the big version of the image, 1 pixel from the mesh on left= ~460m.

I will compute, then post all octants in the next days. I hope this could be useful, especially if Spirit could see Eastern part of the Gusev Rim in the next days or weeks (depending on Spirit route, and on atmospheric dust levels?)

I have finished to render synthetic horizon for North East South far Spirit landscape.
Maybe we will have wonderful pictures of Eastern Gusev's rim, I wonder when I see the results.
I hope Spirit will soon picture Eastern Gusev landscape, but if You want a first synthetic glance, here are two pictures linked:

The first one figure five successive synthetic horizon from North to South, by East (each one: 50 degrees horizontal FOV) . As said before, we cannot expect to see close (ie < 20 km) and/or small lanscape details, but beyond ~20 km, features seems to look fine (see Thyra crater walls from ~45 NE to 90 E).
On this picture, the brown color of the surface fade as a function of distance, to became light grey when >~80km. It seems from current weather on Spirit site that we may not see far landscape as clearly as three months ago.



Then, the second picture figure the mesh used to compute the first one. Color is simple brown, light was approximatively located as the sun will be on a middle afternoon, in winter.

erwan - that is very cool. thanks.

Thanks Lyford, amateur of both mineral and flowered landscapes, you are welcome!

As a variation for the "orbital" view of the mesh, I have mounted on Spirit POV-ray model a very heavy-duty yellow spotlight. Then I think we could more precisely point out the landmarks seen from the Rover:

Beautiful work. here is another view from Mars Express

Erwan,

Amazing -- I'm in awe of your skills.

(And I hope that JPL folks realize what a talented cadre of amateur analysts exist out here.)

Very nice work! that orbital view is very useful!

Alan, Tom, NIX: I am very pleased with your comments, thanks a lot! Pleased especially because of these true color and planet map pictures you share on this forum.

#First: for those (like me?) who like exactness, I have made an error : It's the MOLA/mesh scale, not the picture scale... Anyway, there is a new mesh below.
#Finally: I have made the West part of Gusev crater mesh, and post two images:

- A completed orbital view lightened with Spirit; scale is 293m/pixel.



- A completed octant horizon picture



I will make an annotated version of these pictures next days or weeks, but first try to produce a nice 3D movie from a navcam RNL mesh...
PS : a credit to the site where I send the shared pictures:
Free Image Hosting [xs.to]

Erwan I'm working on a pan of the rim, here's a 'sneak preview'. I'm gonna get some sleep now. I'll finish it tomorrow...
Nice new horizon maps and 360 Gusev! Great.
Cheers

NIX: Good morning! Thanks, wonderful! A was sticked with the image I post now, and don't know that Spirit reveal in part the East Gusev Rim! I'm eager to see Exploratorium, and your finalized Panorama...

-Today I have refined the Gusev mesh: corrected the MOLA cylindrical projection for planetoid, displaced Spirit model 1km North, 0.5km East. Some discrepancies between PANCAM and mesh images vanished.

-I will post the updated (and more correct) orbital view tomorrow, but now I post a "comparison" between a BW L6 PANCAM west panorama from sol 324, and the horizon view from the mesh. It's a 8000x520 0.8MB JPG image:
SPIRIT-sol320-WESTpan-MeshGUSEV

Some nearby landmarks don't appear, but I'm very satisfied with Rim features! The next step is to mark each notable landmark, both on orbital and horizon POV-ray views. So I'll may finally identify each feature seen on Spirit horizon. Thank You MGS-MOLA, MER and POV-ray Teams!

Good morning to you too! I'm scanning the entire set of images for the walls now. I have finished a first version of the panorama for the east wall. If I'm not mistaken a portion of Thyra crater's remaining wall is visible in the foreground just next to Columbia Hills. There is an image of the Hills to be added I think but I will leave that and redo the pan now with 12-bit files. Should be fun...
Lovely horizon view there for sol 324! I'm also gonna try making a 360° pan in B/W of the wall from different sols. Maybe we can overlap our work.
You will have an important piece of work when you add the foreground in your mesh BTW!

Later version with north end of Columbia Hills included.

NIX: I will be very pleased to mix our work. What I may do rapidly is to annotate on a synthetic image the landmarks seen in the color panorama you have made: http://mer.rlproject.com/index.php?act=Att...pe=post&id=5727 ; Then send the picture to you, for a montage or anything else...
To my opinion the view from your Pan extend roughly from NNE to ESE, and effectively we can see a butte who is almost certainly the WNW extremity of Thyra's remaining walls. Is the panorama scanned from the sols near Bonneville pictures?
Dont hesitate to use my E-mail adress if you need.

Great Erwan, do your thing with the pan, I look forward to the results. You mean the sol 68 pictures for Bonneville or all? I'll scan them tonight. Mixing the work would be superb. I'll mail you.

I have linked below a refined version of the orbital view of GUSEV as seen by Spirit.
For those interested, this image is more accurate than the previous version I posted
Differences are evident on SouthWest part of the rim (less features visible). The reason is cylindrical projection of MOLA datas was corrected to spherical projection, taking latitude into account.



MeshGusev1000x800V2.JPG

Sorry, the good link is MeshGusev1000x800V2.JPG

Nix posted here a nice color pan revealing North to East Gusev rim:



Here is pictured our interpretation for this Pan. The two synthetic images show the lanscape seen on NE, F.O.V~80°, from Spirit sol 134 position; the first one from the soil, the second one from 12 km above. Foreground features and Columbia Hills are not represented. The NE part of Gusev Rim is seen entirely and is 80-85 km far (blue cones). The nearer Butte seen beyond foreground, right side of the color pan is a part of Thyra crater rim, 25 km away (green cone). Cone sizes (each cone is 1 km high) are related to distances.
We hope you will enjoy!

Link ot big image (fit with NIX color pan width, 5603x2450, 0.63 MB JPG)
http://xs17.xs.to/pics/05083/MeshGusev-PanNIX-Sol134-01.JPG

Smaller version (1000x525, 70 Kb JPG)
http://xs17.xs.to/pics/05083/MeshGusev-Pan...134-01small.JPG

Ok, now I'm getting the hang of what you're doing...now if you can narrow it down to the immediate surroundings of Columbia Hills so we can see exactly where Spirit is in relation to where it's going...(I don't ask for much, do I?)

MizarKey: Thanks for your post. What you're asking for have to be done, but dealing with the MOLA resolution (463 m /pixel: don't allow to describe Columbia Hills relief) and with my amateur methods, it will need a few days to compute an approximative mesh depicting Columbia Hills topography.
In fact I'm trying to compute such a mesh, starting with an anaglyphic orbital view of Columbia Hills (certainly from JPL, although I'm unable to retrieve the original downloaded file).
Any ideas about downloadle datas adapted to compute such Spirit surroundings topography will be helpful for me!

NIX, Erwan, you guys are doing a tremendous job! The images are awesome!!

Thanks for that aldo! I have been enjoying your website too; it is a nice piece of work!

In the process of "meshing" Columbia Hills, starting from unprocessed raw MGS images, i have made an anaglyph of Husband hill. We may see nicely, i think, Tennessee valley and Larry's lookout relief on the anaglyph. Hope you enjoy!

For those interested, the anaglyph was made (as one previously released) from MGS MOC images E03-00012 (right image) and R02-00357 (left image).
This is interesting to my opinion, because the great majority (if not all?) of MOC image heads toward Nadir (or as least very close), as E03-00012 (see image below). Image R02-00357 (produced later, but before Spirit arrival on Gusev) was certainly from a special attempt to produce elevation data from MOC images, and was the only MOC image i accessed who is heading 18° south from Nadir.
Then the two images, when oriented North on left, could make an anaglyph and allow for elevation datas computing.
MarsExpress use a very similar way, although on a single orbit, to produce elevation datas...

Which software are you using to generate elevation data/meshes from MGS stereo pairs like the one above ?

BTW I assume you have probably seen this page at my website:

http://www.mmedia.is/bjj/3dtest/mars_gusev/index.html

Here I used MOLA data only to generate the DEM used for rendering this so Columbia hills are flat.

Bjorn: thanks for your post. Indeed i have visited your great site, seen planetary maps, Cassini stuff and POV-Ray models you have made!
I don't use any special software to generate elevation datas from MOC images; moreover, alas, i cannot generate correct mesh from these datas for the moment, and posted only one anaglyph from 2 images. The fact is i'm trying to compute exact pixel correlation for any pixels of the two images; with Excel .Then it will be easy to compute elevation for each pixel (though for each squared 20x20 pixels will be more realistic) .It seems that may work...

It would be really cool to compute similar horizon views for some areas without landers...drape some MOC images over the top, and see what one gets. It would be a lot like an MEX image, I imagine, but higher resolution. As more and more MOC repeat coverage exists, I imagine this could be done.

Share my fantasy; by 2020 or sooner a virtual software utility based on super resolution complete Mars visual-geological-hydrological-topographical coverage that lets you explore Mars in a Flight Simulator environment.

you could choose the overlap; visual-geolo...and maybe even land and still have good resolution thanks to MRO-resolution-like imagery.

www.polygonworlds.com

I made the little airplane

Doug

X-Plane has Mars using MOLA data - no bitmaps though, but the demo is free and includes a bit of Mars for your pleasure. You can even simulate Opportunity's EDL... Demo works with joystick for 6 minutes, but you can still use auto pilot after that...

Thanks, i will visit X plane and polygonworlds; i'm fed up with Excel and meshes!

As a suite for the Columbia Hills anaglyph, here is a paralell view of Husband Hill, Tennessee Valley and West Spur; though the relief is exagerated, no doubt Spirit is a good climber, to reach this tiny Larry's Lookout peak!
A brighter patch of soil is seen on the upside Tennessee valley near Husband Hill summit. Present on the left side image (R02: 2003), but not (or barely visible) on the right side image (L03: 2001). Not noticed anything special on Spirit images though... Maybe only an artifact from illumination conditions? Streaks and albedo variations blurr the relief sensation on the plain...
I will notice as well the smooth depression on West Spur base; maybe analogous to transition terrain zones seen on many MOC images of Gusev crater, between flat plains, and hills, mesas, big craters?

Hope you will enjoy...

I have almost finished an Husband Hill mesh. Need some refinements, but a first rendering is linked below (JPG 1000x750, 77Kb).
Spirit path from sol 150 is approximatively figured, and a tiny model (on scale!) of Spirit is located on Larry's Lookout (magnified 3x on insert). A trek!

Awesome! Congratulations on a great job!

Ahh - now that IS amazing - is there any way you could sort out a .wrl or a .3ds or a any other 3d model format of that?

The climb up west spur and around to Larrys Lookout is a very difficult traverse to explain in a lecture with just pictures - and if I could use a 3D model and put the MOC imagery over the top, then I could have a flyaround that shows it much MUCH clearer!

Doug

Doug: thanks for your post. I may post or send to interested fellows a RAW triangle text file, or a POV-ray compatible mesh text file, say in 2 weeks, for Columbia hills entire area (and corrected for biggest errors). Meanwhile, i may render one or few images or a small sequence of images if you need a specific one (and if i could!)

Erwan, looking good! I can't wait for your first 3d rendition/movie........The stereo pair of Tennessee Valley you posted has a lot of good detail. Any chance of you of generating a corresponding stereo pair that would cover the terrain immediately to the east of your posted stereo pair. Specifically, I'd be interested in the black feature southeast of Husband Hill.

Thanks,
Aldo.

Aldo12xu: thanks for your post; linked below is the parallel view you requested, i think; hope you will enjoy!

Wow, that's really cool!

Ummm... I'll start the 3d project of the immediate Opportunity area. It'll be just one flat polygon...

erwan, I used POVRay about a decade ago, it's one cool program. I've since tinkered around with trueSpace every once in a while... If you haven't already, an older version of Caligari trueSpace (ver 3.2) is now completely freeware, you can download it here. It will ask you to register but that's pretty much it, they will email you a serial number... tS3 doesn't quite match up with the latest version they have, or any other modern 3d modeling software but once you master the funky interface it can do some nice scenes...

I remember they were worried before landing that Meridiani would be TOO flat and thus boring!
Now on to the etched terrain!

erwan - I am in awe of your mad POV skills.... I haven't used that program since (gasp) my Atari days....! been so caught up in GUIs that I forgot how powerful that engine was - great work.

I remember I did a project for school back then (1993 or so), and set up a render farm from a couple of 486 machines which cranked away over the weekend to produce a few-hundred frame animation. Full raytracing and all, using only POV-ray engine with RTAG animation generator. One of these days I'll dig up the original frames and put it back together again with better aliasing...

Looking at it now, it appears to almost perfectly match the Columbia Hills region, complete with a swimming pool

Swimming pool at Gusev

Yes, IIRC, my experience with POV usually involved a mirror ball, a checkered floor, a transparent (!) glass chalice and then maybe the USS Enterprise. Being a po' musician, I owned an Atari ST and all my graphics friends had Amigas or Macs.... State of the Art, you know.

BTW, Nice water, Pando - though I don't think it would last on Gusev... Too cold to swim anyway.

Found this acticle, it may be of interest...

USGS HIGH-RESOLUTION TOPOMAPPING OF MARS WITH MARS ORBITER CAMERA NARROW-ANGLE IMAGES
IAPRS, Vol. 34, Part 4, “GeoSpatial Theory, Processing and Applications”, Ottawa, 2002

http://webgis.wr.usgs.gov/mer/March_2002_p...02_Kirk_MOC.pdf

More here:
http://webgis.wr.usgs.gov/mer/moc_na_topography.htm

...and even more here (including a 75MB one page PDF file about the geological features in the Gusev region, you need a beefy computer just to open the darn thing!):
http://pubs.usgs.gov/imap/i2666/

Pando: thanks for your post; your movie is the last nice evidence of water on Mars! A rapid glance at the article you posted; i will (try) to read the paper thoroughly later... I hope i will find tips and tricks to speed up a very tedious Excel process!

OK- so how do we GET these meshes - they'd make cool VRML thingies

Doug

Pando; I have a zoom in of that USGS geological map on my 'baby-website'; it is in a 'testing phase' and I'm slowly uploading processed imagery. Choose the 'Spirit and Opportunity location' link. I would also like to hear how my pages are viewed; I'm at the beginning of a big upload and tips or remarks could be useful.

NIX Mars Mission Imagery