AUTOMATIC ROAD EXTRACTION BY INTEGRATED ANALYSIS OF IMAGES AND GIS DATA

Accurate 3D road network is a vital component of GIS for many applications, including traffic management, monitoring, city modeling, and visualization. This paper presents a practical system for automated 3D road network reconstruction by integrated processing of color image data and information from existing digital spatial databases. Starting from the existing database, the road attributes and the approximated road geometry are derived. Guided by this knowledge, various features including 3D straight edges, road regions, shadows, roadmarks and zebra crossings are extracted by the developed algorithms. The system then uses and fuses these features, existing information to generate and group road primitives to extract the roads. The key of the system is the use of knowledge as much as possible to increase success rate and reliability of the results, working in 2D images and 3D object space, and use of 2D and 3D interaction when needed. Another advantage of the developed system is that it can correctly and reliably handle problematic areas caused by shadows and occlusions. To complete the road network reconstruction, an efficient approach for road junction modeling has also been developed. The system was originally developed to process stereo images, but it has been modified to work also with orthoimages, thus making it applicable to sensors of unknown geometry. The system has been implemented as a stand-alone software package, and has been tested on a large number of images with different landscape. In this paper, various parts of the developed system are discussed, and the results of our system in the tests conducted independently by our project partner are presented together with the system performance evaluation.