Off-line constrained vocabulary cursive script recognition using visible features

This paper presents a model for off-line cursive script recognition. The method proposed combines both analytical and holistic approaches to solve the problem of cursive script recognition. The emphasis is to create a fast and reliable model for recognition. The holistic approach of extracting feature is used with the analytical approach of segmenting and recognizing the first character. Pre-processing, feature extraction, classifier, and phrase recognition are explained and used in this system. Results from a test set of 1294 images are presented based on three different word recognition methods that are experimented. This system is being used as a system to sort mails that are directed overseas, however, it can also be used for other requirements like word spotting in unconstrained text.

[1]  Graham Leedham,et al.  Empirical design of a holistic verifier for automatic sorting of handwritten Singapore postal addresses , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[2]  Sargur N. Srihari,et al.  Parsing and recognition of city, state, and ZIP codes in handwritten addresses , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[3]  Ching Y. Suen,et al.  Automatic reading of cursive scripts using human knowledge , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[4]  Ching Y. Suen,et al.  Cursive script recognition applied to the processing of bank cheques , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.