Matching of road data with different scales for updating

GIS data can be updated using the larger scale GIS data based on map generalization. Matching is a critical first step for extracting updates in this process. This paper presents an approach for road data matching based on level analysis. Matches of objects between road networks are divided into three levels, i.e. decomposed level, basic level and abstracted level. The matching of edge end point as the decomposed level, the node matching as the basic level, and the route matching and between node and edge matching as the abstracted level are discovered. Based on the order relationship of three matching levels and the interdependent among them, matches of objects are accomplished by a set of algorithms developed. Meanwhile during this process, the exceptional matching can be as the index for manual checking to improve the matching accuracy. Matches of objects meet the requirement for searching data unmatched. The experiment results of matching road data at 1:10,000 and 1:50,000 show higher matching accuracy.