Offline recognition of large vocabulary cursive handwritten text

This paper presents a system for the offline recognitionof cursive handwritten lines of text. The system is based oncontinuous density HMMs and Statistical Language Models.The system recognizes data produced by a single writer.No a-priori knowledge is used about the content of the textto be recognized. Changes in the experimental setup withrespect to the recognition of single words are highlighted.The results show a recognition rate of ~85% with a lexiconcontaining 50'000 words. The experiments were performedover a publicly available database.

[1]  Mounim A. El-Yacoubi,et al.  Conjoined location and recognition of street names within a postal address delivery line , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[2]  Proceedings Seventh International Conference on Document Analysis and Recognition , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[3]  Horst Bunke,et al.  Using a Statistical Language Model to Improve the Performance of an HMM-Based Cursive Handwriting Recognition System , 2001, Int. J. Pattern Recognit. Artif. Intell..

[4]  Juergen Luettin,et al.  A new normalization technique for cursive handwritten words , 2001, Pattern Recognit. Lett..

[5]  Mark Liberman,et al.  THE TDT-2 TEXT AND SPEECH CORPUS , 1999 .

[6]  Frederick Jelinek,et al.  Statistical methods for speech recognition , 1997 .

[7]  Anthony J. Robinson,et al.  An Off-Line Cursive Handwriting Recognition System , 1998, IEEE Trans. Pattern Anal. Mach. Intell..