Recognizing rectangular objects in digitized topographic maps

An object-oriented approach to unsupervised vectorization of rectangular objects in topographic maps based on their recognition by geometric features is presented. An implementation of this approach led to a two- to threefold reduction in labor needed to develop vector models of buildings and industrial installations from their maps. Andrei V. Chernov. Born 1975. Graduated from the Samara State Aerospace University (SSAU) in 1998. In 2004 defended a Candidate dissertation (Eng.). Currently a senior lecturer at the Chair of Geoinformation Systems, SSAU.Scientific interests: geographical information systems, processing of remote sensing data, and pattern recognition. Author of more than 50 publications, including 24 papers in scientific journals, and a co-author of a monograph. Member of the Russian Federation Association of Pattern Recognition and Image Analysis. Nikolai V. Chupshev. Born 1986. Graduated from the Samara State Aerospace University (SSAU) in 2008. Scientific interests: image analysis, geographical information systems, and pattern recognition. Author of three publications including one journal paper.