A new Voronoi diagram-based approach for matching multi-scale road networks

Abstract Object matching is a key technology for map conflation, data updating, and quality assessment. This article proposed new Voronoi diagram-based approach multi-scale road networks (VAMRN). Using this method, we first created diagrams of the network using strategy discretizing lines into points adding dense to special intersection segments. Then, used diagram segment find candidates. Finally, obtained results by judging geometric similarity metrics designed heuristic combination optimization strategy. The experimental demonstrated that VAMRN outperformed two existing methods in generality quality. F-measures were 18.4, 29.6, 3.8, 7.6% higher than buffer growing 4.5, 2.8, 1.8, 6.1% probabilistic relaxation method. And time performance improved more 90% over