Geographical Electronic Map Object Recognition Research

Digital geographic information is the basis of Geography Information System (GIS), and to capture the information is one the keys to build and develop GIS. To recognize and capture digital geographic information in the scanned maps images has transcendent advantages, and has become the main method to capture the digital information.After studying the nowadays methods, combining the characters of binary scanned maps images, the author designs a method to recognize and capture the digital geography information, which is based on thinning algorithm. A system is also realized by this method in the Visual C++ programming environment.Vectorization method based on thinning algorithm used in the design prevails currently, whose progress is: after the binary scanned map images are thinned, recognize the skeleton captured. During the progress, the result after thinning affects the final effect of vectorization directly. The author designed a thinning algorithm combining 3x3 template with 4X4 templates, which keeps the topology structure of the primary maps successfully. In order to recognize and capture figures, vector exterior product is combined with minimal-angle principle, so this method can get polygons more rapidly. Otherwise, the author applies vector-searching algorithm, lines-recognizing algorithm, noise -removing algorithm and some basic operation towards binary images. The System built can get an excellent result after vectorization.