3d Terrain models

[ajf]

I've been researching several forums looking for software that will take DEM files (Digital Elevation) and allow you to convert them over to a 3D model file format that you can use in Aspire to create 3D contour map. I think I've settled on the using the program AccuTrans 3D http://www.micromouse.ca/index.html to do this. What I'm wondering is if there are any other users of AccunTrans out there that might be willing to share some tips, tricks or insights into this program that might save me some learning curve time.

[jerry_stanek]

could you take a topical map and use the 1 rail sweep to do ihat

[myxpykalix]

Al,
I haven't used accutrans in a while but if i recall correctly it does file conversion mostly so you should be able to load a dem file and have it spit out a stl for input to whatever pgm you want to use.

I used it last time to do a conversion of Monticello here in Va. to a stl that i then carved.
Other then file conversion what are you trying to do with it?:confused:

[cnc_works]

It has been a couple years, but I have used Accutrans successfully to produce stl files for cutting just by following the basic instructions.

My issues were finding appropriate dem files for conversion, especially for smaller areas, the Portland metro area for example. Or a local lake a few miles long. Finding the right scale covering the necessary area was always problematic, so any hints on this aspect of 3D contour files would be appreciated.

[ajf]

I want to create a 3D topo map of the region I live in. Using the DEM files from GeoBase Canada for our region and bringing them into AccuTrans and just doing a straight conversion to STL and then importing that into Aspire ends up giving me a model with no visible vertical detail in it. So I know I'm missing something just not sure what it is right now. It might be something obvious in AccuTrans that I'm not seeing, but I've spent a bit of time so far and haven't found the answer.
What I have found is if I take the DEM file and output it as grey scale BMP file and then do the conversion of the BMP to a component model in Aspire. I end up getting a reasonable contour map of the region. But I'm assuming that STL or 3DS file should give me better detail.

[dana_swift]

The best source of usable DEM's I have found for the US is:
http://www.webgis.com/terr_dem75_data.html

That doesn't help with canada data tho. Once upon a time I had a link for a world DEM database of lower resolution, it may still exist. That would be useful for anybody.

There are several kinds of DEM's, its a very technical file format. One kind of DEM is a grid of elevation points on north-south, east-weat grid lines. That kind of DEM is easy to read but has the problem as you go north, there is distortion from the spreading of the north-south lines. It can be scaled by the cosine of the latitude to get a very reasonable image. It would not stitch together to make a globe because the northern border would be the same width as the southern border. The globe makes the northern border shorter than the southern one in the north of the equator.

This kind makes a good "picture frame-able" contour plot of an area.

So .. there is another kind of DEM where all that is exactly geoid correct, but it is a mathematical mess to read them. They stitch together to make a correct plot of the earth correctly, but they are messy to cut as the edges are geoid projections.

The earth has this nasty property of being about a thousand miles narrower north to south through the poles, than it is wide at the equator. Due to our spinning and centrifugal forces.

My suggestion is to try several kinds of DEM and see what you like better for your application. The bad news is they are not marked in simple English, instead they are labeled as "ortho projections", "Mercator projections", etc. Then according to the latitude angle included in the data. Its not easy to tell which kind you have. Some are in meters, some are in feet, just to make it more fun. No information about that on the file name unfortunately. The software reading it has to know the difference and correct the units to your preference.

Usually a DEM repository has files in a similar format, so when you find one.. bookmark it! Hope the data at the link is helpful.

Thought you might find the commentary useful. If not.. ignore.

D

[dana_swift]

Addendum-

I just noticed the same database I provided the link for has the "geotopo" database of DEM's for the whole world. Yet another brain cell that didn't survive intact..

http://www.webgis.com/terr_world.html

They do note that vertical units are meters, and the horizontal resolution is just under 900m or about a quarter mile.

Al- your comment about having no vertical relief is what you would expect without special scaling. If the X and Y are scaled in meters, the Z relief is such a small percentage it is virtually flat. If the earth was scaled down to the size of an orange, the orange would have a much rougher surface.

D

[billp]

For years people were using the program called "Wilbur"to aid in translating DEM's, etc. They still have a relatively current version posted on the web, and I'll include some links below.
There is also another program called Geofrac 2000, and they claim to make Wilbur obsolete...Not sure which is the best for your current applications, but they both might be worth a peek...
http://www.rhttp://www.ridgenet.net/~jslayton/wilbur.html (http://www.rhttp://www.ridgenet.net/%7Ejslayton/wilbur.html)
http://idgenet.net/~jslayton/FunWithWilburVol1/index.html (http://idgenet.net/%7Ejslayton/FunWithWilburVol1/index.html)
http://www.ridgenet.net/~jslayton/wilburchanges.html (http://www.ridgenet.net/%7Ejslayton/wilburchanges.html)
http://www.ofpec.com/forum/index.php?topic=3212.0;wap2

[ajf]

Thanks for the useful information. I know I'll eventually understand it and probably enjoy the challenge to make this work. I'll first make sure I understand the DEM format that is available at the Canadian sites, and go from there.

[cnc_works]

Just for the heck of it, I downloaded the 30day test version of Accutrans, opened a 30m DDF file of Portland, saved as STL, then opened up the stl in Rhino for a perfect 3D contour file.

Here are the two files I used for test purposes.

http://dl.dropbox.com/u/20960044/3dcontourtest.zip (http://dl.dropbox.com/u/20960044/portland30metermap.DDF)

[tomwillis]

Al - what I've been doing is using Google Sketchup and an STL export....all free. The detail is decent but not like DEM files. Feel free to contact me - and I can walk you thru it. Fairly simple.
Tom

[ajf]

I found by using the programs Microdem and AccuTrans 3d I've been able to manipulate DEM files in such a manner, that when I bring them into Aspire as a STL file they give me the level of detail that I was looking. Still a lot of things to learn about DEM files though.

http://www.usna.edu/Users/oceano/pguth/website/microdem/microdem.htm

[chiloquinruss]

I played around with it for a few minutes and came up with this for the sake of further discussions.

http://hobby-tronics.lenzus.com/content/21775739/r/s_47947755.jpg

The topo pdf had no vectors, so I created a jpg, then inverted it so that black was white and white was black, I then opened Aspire with a 4 inch by 48 x 48 piece, and imported the inverted bitmap, stretched it to fit, then converted the bitmap to a 3d component, I then set the max depth of the compnonent to 4 inches, then I toolpathed a 'final' 3d pass with a 1/4 inch ballnose bit selected. It's certainly not a finished product but I think it may let you do some experimenting on your own. If you are cutting foam you can really fly and may not need to do a roughing pass. Anyhow I was just playing around and this is what I came up with. Have fun! :D Russ

[ajf]

I'm finding this project quite interesting and I'm probably spending far to much time on it. But it is nice to get the old mind researching and learning new things, sort of like being back in university again. I came across a free open source GIS program that I found just amazing in what you can do with it, called the Quantum GIS Project, if anyone is interested. http://www.qgis.org/