Combination of Invariant Pattern Recognition Primitives on Technical Documents

This paper deals with a particular aspect of a technical document interpretation device: the recognition of multi-oriented and multiscaled characters and symbols. The adopted methodology is based on original descriptors, relying on the computation of the Mellin Fourier Transform. These descriptors are then combined with classical invariant through the use of Genetic Algorithms. The application frame of this study is the interpretation of the documents of the French telephonic operator France-Telecom for which the recognition of symbols constitutes a crucial point for a robust interpretation.

[1]  Anil K. Jain,et al.  Feature extraction methods for character recognition-A survey , 1996, Pattern Recognit..

[2]  J. Ogier,et al.  Contribution à l'analyse automatique de documents cartographiques. Interprétation de données cadastrales , 1994 .

[3]  Yves Lecourtier,et al.  Multi-scaled and multi oriented character recognition: an original strategy , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[4]  Jack Sklansky,et al.  A note on genetic algorithms for large-scale feature selection , 1989, Pattern Recognit. Lett..

[5]  Anil K. Jain,et al.  Algorithms for feature selection: An evaluation , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[6]  Mohan M. Trivedi,et al.  Circular-Mellin features for texture segmentation , 1995, IEEE Transactions on Image Processing.

[7]  Majid Ahmadi,et al.  Pattern recognition with moment invariants: A comparative study and new results , 1991, Pattern Recognit..

[8]  Faouzi Ghorbel,et al.  A complete invariant description for gray-level images by the harmonic analysis approach , 1994, Pattern Recognit. Lett..

[9]  Robert B. McGhee,et al.  Aircraft Identification by Moment Invariants , 1977, IEEE Transactions on Computers.

[10]  M. Teague.  Image analysis via the general theory of moments , 1980 .

[11]  Alireza Khotanzad,et al.  Rotation invariant image recognition using features selected via a systematic method , 1990, Pattern Recognit..

[12]  N. Kita.  Object locating based on concentric circular description , 1992, [1992] Proceedings. 11th IAPR International Conference on Pattern Recognition.