A basin recognition method by landform classification and geometrical feature discrimination

Aiming at the geomorphological and geometrical features of basin, a recognition approach combined with landform classification discriminant analysis is presented. First, k-means algorithm applied to digital elevation model data. The basic types landforms elevation, slope, relief amplitude are clustered, regions plains, mountains, hills have been separated. Focusing on plain its neighboring region, measurement function defined analyze cases surrounded by mountains hills. Finally, discrimination for basin presented parameters state, area plain, difference elevation. DEM Shuttle Radar Topography Mission China. experiments show that results same as those real terrains.