Over the last year or so, I made a number of posts to this forum regarding a project I was working on and received a lot of good advice. Well, I’ve finally got my stuff to the point where I felt it was ready to be shared. It is now available on github at https://github.com/gwlucastrig/Tinfour.
Tinfour is a Java library that provides construction and management of 2D Triangulated Irregular Networks (TINs). TINs are useful in the analysis and rendering of surfaces. While 2D triangular mesh structures are most often applied to modeling terrain and elevation data, they can be applied to numerous applications.
Anyway, Tinfour came out pretty well and runs pretty fast. It can process more than a million samples per second and requires only 124 bytes per vertex to represent a TIN. I’ve also put together a simple test application for rendering data from elevation data collected using airborne Lidar (laser measuring devices). Lidar is a really cool technology and the data it produces is downright amazing. If you want to try using Tinfour to view some of it, I recommend starting with tile 41001330PAS from the PAMAP collection. At 36.7 megabytes, it isn’t dainty, but it does contain some really interesting land features.
The github project is brand new and I don’t have a lot of experience deploying software or setting up projects on github… So if anyone might be interested in this kind of thing, I’d welcome any suggestions or insights you might be able to offer. Likewise if you’d care to comment about the software itself.
"The good news about computers is that they do what you tell them to do. The bad news is that they do what you tell them to do." -- Ted Nelson
Gary W. Lucas
posted 3 years ago
Thanks for the kind words and cow recommendation.
If you are viewing the data, you might want to try going into the View -> Styling and Presentation menu and turning on the hillshade option. This option uses an illumination model to add shading to the imagery and gives a better sense of the surface. The data set I recommended is relatively small (1.5 million sample points), so the "Build Grid with Full Resolution" option which will work just fine if you want to see maximum detail.
The PAMAP data I recommended is from a 2006 aerial survey and gives surface elevation measurements with an 8.7 foot average spacing. More recent surveys are much denser, often with a "nominal point spacing" of 0.3 meters or better.
Incidentally, the PAMAP site has a general introduction to Lidar and Pennsylvania's collection at PAMAP Lidar Elevation Data. Lidar is a relatively new technology and the Pennsylvania survey was conducted "early in the game." Even so, the PAMAP folks managed to do the job right... better, I think, than some of the folks who have come after them. So if you're interested in terrain modeling, the site is a valuable resource.