ONLINE RECOGNITION BY DEVIATION-EXPANSION MODEL AND A-ASTERISK ALGORITHM-BASED MATCHING

This paper presents an online recognition system for large-alphabet ha ndprinted Chinese characters using a model-based recognition approach with stroke-based features. A deviation-expansion (D-E) model represen ting the reference pattern is constructed. The model contains hypothet ical knowledge of handwriting variations, including stroke-order devia tions and stroke-number deviations. For pattern matching, a matching t ree is constructed by combining the knowledge of the reference pattern and the unknown pattern together. With the tree a similarity measure function is defined to indicate the degree of similarity. Evaluation o f the function is obtained using A algorithm-based matching. Experime ntal results are based upon testing a set Of 54010 handprinted sample characters written in the square style by ten people. The cumulative c lassification rate of choosing the ten most similar characters is 98%. The results suggest that the hypothetical model is both feasible and reasonable.