Reconnaissance de courriers manuscrits par HMM contextuels et modèle de langage

This article describes a global approach for the recognition of handwritten lines which was submitted at the ICDAR 2011 French handwritten mail recognition competition. Its main contribution is to extend the context-dependent HMMs word recognition system to the line level without segmentation into words. Thus, the authors describes a complete solution for mail recognition: original preprocessing, context-dependent HMMs and an optimized bigram language model. Preprocessing reduces word error rate by 5% in absolute value.