INTEGRATING STATISTICAL AND STRUCTURAL APPROACHES TO HANDPRINTED CHINESE CHARACTER-RECOGNITION

Handprinted Chinese character recognition (HCCR) can be classified int o two major approaches: statistical and structural. While neither of t hese two approaches can lead to a total and practical solution for HCC R, integrating them to take advantages of both seems to be a promising and obviously feasible approach. But, how to integrate them would be a big issue. In this paper, we propose an integrated HCCR system. The system starts from a statistical phase. This phase uses line-density-d istribution-based features extracted after nonlinear normalization to guarantee that different writing variations of the same character have similar feature vectors. It removes accurately and efficiently the im possible candidates and results in a final candidate set. Then follows the structural phase, which inherits the line segments used in the st atistical phase and extracts a set of stroke substructures as features . These features are used to discriminate the similar characters in th e final candidate set and hence improve the recognition rate. Tested b y using a large set of characters in a handprinted Chinese character d atabase, the proposed HCCR system is robust and can achieve 96 percent accuracy for characters in the first 100 variations of the database.