Research on Data Generalization of Land-use Types Caused by Type Up-scaling
Land-use type data at different scales expresses different content, transmits different information, and reveals different phenomena and laws. The macro-data at large-scale is usually generalized from many fine data at small-scale. The process of land-use type data generalization caused by type up-scaling was addressed and the methods of land-use type aggregating, merging of adjacent parcels with the same type, small parcels amalgamating, spatial-data simplifying and topology preserving are elaborated respectively. This paper focused in the approach of small polygon amalgamation by skeleton subdivision, which first used constrained Delaunay method to build triangu-lation network and used the triangulations to build the skeleton lines according to five different types of triangulation and then divided the small polygon using the skeleton lines into many smaller polygons adjacent to different polygons and merged them to the different adjacent polygons. Uses this approach to amalgamate the small parcels which their areas are smaller than criterion, the proportion of different land-use types was roughly the same as before a-malgamation; then proposed a polygon simplification method to simplify land-use type polygons, which was based on the Douglas-Peucker algorithm and constructed a balanced line to preserve the same area as before simplification. In order to maintain the topology of the polygons, first divided the polygon into different polylines according to their adjacent relations and simplified them respectively and then rebuilt the polygon with the simplified polylines. The area preserved simplification process first used the Douglas-Peucker algorithm to simplify the polyline and got the smallest simplification unit and within the unit built a balance line to preserve the same area as before simplification. Thus the area and the topology before and after simplification are consistent. The results of the test data showed that the method is suitable for the generalization of land-use type data, and can be used else where when geospatial data were up-scaled.
up-scalingdata generalizationpolygon amalgamationpolygon simplificationland-use type
NETL
NSTL
万方数据
维普
