High Performance Unconstrained Word

In this paper we present a system for the recognition of handwritten words on literal cheque amounts which advantageously combine hmms and Markov random elds (mrfs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random eld realizations. The hmm analyzes the image along the horizontal writing direction, in a speciic state observation probability given by the column product of causal mrf-like pixel conditional probabilities. Aspects concerning deenition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the srtp 1 French postal cheque database (7031 words, 1779 amounts, diierent scriptors).