Global Topology Concept Neighborhood Graph and Its Application in Topological Similarity Measurement

Abstract

A topological relationship is a key factor in the spatial data processing. Aiming at the problem of topological similarity measurement when the dimension of spatial objects changes in the process of map generalization, this paper proposes a topological similarity measurement method based on the global topological concept neighborhood graph. Firstly, all the topological relationships that can be described by the nine intersection model are abstracted, generalized, and classified, and the classified point/point, point/line, point/polygon, line/line, line/polygon, and polygon/polygon local topological concept neighborhood graphs are established. Then, according to the map generalization rules, the mapping between different types of topological concept neighborhood graphs is established, and the global topological concept neighborhood graph is obtained, Finally, the similarity of the topological relationship between different morphological objects is measured. The experimental results show that the proposed method can better meet the topological similarity measurement when the shape of spatial objects changes.