Automatic Bridge Detection in High-Resolution Satellite Images

A set of methodologies and techniques for automatic detection of bridges in pan-chromatic, high-resolution satellite images is presented. These methods rely on (a) radiometric features and neural networks to classify each pixel into several terrain types, and (b) fixed rules to find bridges in this classification. They can be easily extended to other kinds of geographical objects, and integrated with existing techniques using geometric features. The proposed method has been tested in a number of experiments.

[1]  Aude Billard,et al.  From Animals to Animats , 2004 .

[2]  P. Robertson.  GRAVA-a corpus based approach to the interpretation of aerial images , 1999 .

[3]  Josiane Zerubia,et al.  A Markov Point Process for Road Extraction in Remote Sensed Images , 2000 .

[4]  Sei-ichiro Kamata,et al.  A neural network classifier for LANDSAT image data , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol.II. Conference B: Pattern Recognition Methodology and Systems.

[5]  Nicholas R. Jennings,et al.  3D Scientific Data Interpretation Using Cooperating Agents , 1998 .

[6]  J. Desachy.  A knowledge-based system for satellite image interpretation , 1992, [1992] Proceedings. 11th IAPR International Conference on Pattern Recognition.

[7]  Koichiro Deguchi,et al.  An architecture of object recognition system for various images based on multi-agents , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[8]  Alain Ketterlin,et al.  Unsupervised learning of spatial regularities , 1995, Remote Sensing.

[9]  Azriel Rosenfeld,et al.  A Comparative Study of Texture Measures for Terrain Classification , 1975, IEEE Transactions on Systems, Man, and Cybernetics.