OFF-LINE HANDWRITTEN WORD RECOGNITION WITH EXPLICIT CHARACTER JUNCTURE MODELING

In this paper, a new off-line handwritten word recognition method base d on the explicit modeling of character junctures is presented. A hand written word is regarded as a sequence of characters and junctures of four types. Hence both characters and junctures are explicitly modeled . A handwriting system employing hidden Markov models as the main stat istical framework has been developed based on this scheme. An intercon nection network of character and ligature models is constructed to mod el words of indefinite length. This model can ideally describe any for m of hamdwritten words including discretely spaced words, pure cursive words, and unconstrained words of mixed styles. Also presented are ef ficient encoding and decoding schemes suitable for this model. The sys tem has shown encouraging performance with a standard USPS database.