VARIABLE DURATION HIDDEN MARKOV MODEL AND MORPHOLOGICAL SEGMENTATION FOR HANDWRITTEN WORD RECOGNITION

This paper describes a complete system for the recognition of unconstr ained handwritten words using a continuous density variable duration h idden Markov model (CDVDHMM). First, a new segmentation algorithm base d on mathematical morphology is developed to translate the 2-D image i nto a 1-D sequence of subcharacter symbols. This sequence of symbols i s modeled by the CDVDHMM. Thirty-five features are selected to represe nt the character symbols in the feature space. Generally, there are tw o information sources associated with written text-the shape informati on and the linguistic knowledge. While the shape information of each c haracter symbol is modeled as a mixture Gaussian distribution, the lin guistic knowledge, i.e., constraint, is modeled as a Markov chain. The variable duration state is used to take care of the segmentation ambi guity among the consecutive characters. A modified Viterbi algorithm, which provides l globally best paths, is adapted to VDHMM by incorpora ting the duration probabilities for the variable duration state sequen ce. The general string editing method is used at the postprocessing st age. The detailed experiments are carried out for two postal applicati ons; and successful recognition results are reported.