On-line free handwritten Chinese character recognition method based on component cascaded HMMs

This paper presents a cascaded hidden Markov model (HMM), which allows states transition, skip and duration. The cascaded HMM extends the way of HMM pattern description of handwritten Chinese character (HCC) and depicts the behavior of handwritten curve more reliably in terms of the statistic probability. Hence the character segmentation and labeling are unnecessary, Viterbi algorithm is integrated with the cascaded HMM after the whole sample sequence of a HCC is input. Compared with the segment HMMs approach, the recognition rate of this model for the first candidate is 87.89% and the error rate could be reduced by 12.4%.