Abstract: Offline Handwritten Chinese Character Recognition(HCCR) is one of the most difficult problems in pattern recognition, and various approaches are presented to recognize their different aspects. This paper describes the main techniques which mainly influence the performance of the recognition system, including the way of normalization, feature extraction, classification and recognition. It evaluates each technique and gives its conditions for correctly use, introduces the currently used Chinese character database.

Key words: Handwritten Chinese Character Recognition(HCCR), normalization, feature extraction, classification and recognition

摘要： 脱机手写体汉字识别是模式识别领域中的难题之一。该文分析影响脱机手写体汉字识别性能的主要方面，如规范化方法、特征提取方法及分类方法，给出了每种方法的适用条件，介绍了目前研究中所使用的数据库。

关键词: 手写体汉字识别, 规范化, 特征提取, 分类识别

CLC Number: 

• TP391.43