The Collaborative Simplification Method for Road and Residential Area Considering Shape Similarity
Spatial conflict processing is a difficult problem in cartographic generalization.At present,the focus of spatial conflict processing is topological consistency,and less attention is paid to the shape similarity.Shape similarity is the specific embodiment of spatial correlation in local areas.Road and residential area are a pair of geographical features with strong correlation.After selection,merger,exaggeration,segmentation,and other operations,some residential areas are shown as block with large area.The distance between block residential areas and adjacent roads is small and the shape is similar.At this time,if the roads and residential areas are sim-plified separately,uncontrollable topology conflict and shape conflict are likely to occur.To maintain topology consistency and shape similarity,the paper proposes a collaborative simplification method for roads and residen-tial areas.First,Delaunay triangulation is used to build the adjacent relationship between roads and residential ar-eas and extract the paired adjacent pieces.Then,the dynamic time warping algorithm is used for rough matching of nodes,and further optimization is made according to the included angle of nodes,the angular bisector direc-tion of nodes,and the direction of connecting lines.Finally,the roads are simplified by the algorithm based on node selection,and the result of node selection is synchronized to the contours of the residential areas.After sim-plification,abnormal line segments are detected and corrected.The experiment is carried out on the roads and residential areas in a 1∶500 00 map of a region in Zhejiang Province,and the results show that the method can realize the synchronous simplification of roads and adjacent residential areas on both sides.After simplification,the shape and structure of roads and residential areas coincide well,effectively maintaining the shape similarity,topology consistency,and visual effect of roads and residential areas in adjacent areas.
NETL
NSTL
万方数据
