A TIN compression method using Delaunay triangulation

This study introduces a new Triangulated Irregular Network(TIN) compression method and a progressive visualization technique using Delaunay triangulation. The compression strategy is based on the assumption that most triangulated 2.5-dimensional terrains are very similar to their Delaunay triangulation. Therefore, the compression algorithm only needs to maintain a few edges that are not included in the Delaunay edges. An efficient encoding method is presented for the set of edges by using vertex reordering and a general bracketing method. In experiments, the compression method examined several sets of TIN data with various resolutions, which were generated by five typical terrain simplification algorithms. By exploiting the results, the connecting structures of common terrain data are compressed to 0.17 bits per vertex on average, which is superior to the results of previous methods. The results are shown by a progressive visualization method for web-based GIS.