Update of road network in GIS by fusing multi-sensor imagery

Abstract

Remote sensing, as a tool to obtain the spatial data, is becoming more and more important in the update of spatial data, and remote sensing data are becoming a main data source to update spatial data. A method is proposed in this paper to update road network in GIS automatically by use of the remote sensing imagery. Because of the redundant and complementary information existing in different sensor data from the same areas, which can reduce the uncertainty and improve the accuracy of the features extracted from multi-sensor data, an algorithm is developed to extract road network by fusing radar images and optical images. Then the changes between the road network extracted from different remote sensing images and the roads in GIS database are detected by a change detection algorithm of linear features. Taking each road feature in the road network layer of GIS database as a unit to detect changes, the segments of road features are matched with the corresponding segments extracted from the remote sensing images to conform whether to correct the existing data or to add new road features to the GIS database. The experimental result shows that the method can update the road network in GIS database exactly. © 2005 IEEE.