An efficient reverse geocoding method based on Global Subdivision Model

Abstract

As a critical component of Location-Based Service(LBS), reverse geocoding is the process of converting a coordinate obtained by GPS to a readable street address. With the widespread use of location-aware mobile electronic devices and the extensive application of LBS, the capability of reverse geocoding is in high demand. Meanwhile, with the constantly expanding of spatial data size, the computing quantity of reverse geocoding continues to increase. Based on Global Subdivision Model(GSM), this paper presents an efficient reverse geocoding method named Geohash-based reverse geocoding(GBRG). In GBRG, spatial objects are encoded to Geohash codes which can be used to identify the location of the spatial objects. A key point in this step is encoding spatial line objects and region objects with an adaptive Geohash method. Then the spatial objects near the given coordinate are retrieved by comparing the Geohash codes. In order to make sure that the nearest spatial objects are in the scope of the results retrieved, we use the eight neighborhood for retrieval. PostGIS provides a lot of functions for geometric calculation and we use these functions to determine the spatial relations between the given coordinate and the spatial objects. We carry out some experiments using the data of Beijing, China. The results show that GBRG is of high efficiency with some loss in accuracy which can be ignored without obvious influences.