The TerraScan Ground Routine is designed to classify a ground surface from point data. There are multiple parameters within the filter. The following chart is a summary and general guidelines for setting the parameters.
|Max Building Size||Grid size for initial seed surface||Smaller is more aggressive, higher fidelity, but more false ground points in initial surface|
|Terrain Angle||Max terrain angle||Maximum slope plus 10 – 15°; 88° in areas of near vertical features (cliffs, retaining walls)|
|Iteration Angle||Maximum value at which a point will be added to the ground surface||Higher is more aggressive; typical is 4 – 10°|
|Iteration Distance||Maximum height above surface at which a point will be added to the ground surface||Higher is more aggressive; typical is 0.5 – 1.5 m|
|Reduce Iteration Angle||Soft stop; adds less points in areas with greater point density||½ of desired grid spacing|
|Stop Triangulation||Hard stop; ensures no triangles with an edge smaller than the specified value||¼ of desired grid spacing|
The TerraScan Ground Routine algorithm is implemented based on the widely used Axelsson algorithm.
Axelsson, P., 2000, “DEM GENERATION FROM LASER SCANNER DATA USING ADAPTIVE TIN MODELS”, International Archives of Photogrammetry and Remote Sensing, Vol. 33, part B4/1, ISPRS, Amsterdam 2000, pp. 110–117.