用于手写体汉字识别的汉字结构模型  

Chinese Character Structure Models for Handwritten Chinese Character Recognition

作　　者：刘峡壁[1] 贾云得[1] 

机构地区：[1]北京理工大学信息科学技术学院计算机科学工程系,北京100081

出　　处：《北京理工大学学报》2003年第3期322-326,共5页Transactions of Beijing Institute of Technology

摘　　要：针对手写体汉字识别问题,选取笔段和笔划作为基元,分析手写体汉字的组成规律和变形规律,提出了两种汉字结构模型:笔段中心点模型和笔划关系矩阵模型,以及基于模型的分类依据和识别方法.根据所提出的模型,采用两级分类方案构造汉字识别系统,粗分类采用笔段中心点法,细分类采用笔划关系矩阵法.实验表明模型是有效可行的.By choosing stroke and substroke as basic components to analyse the composition regularities and deformation regularities of handwritten Chinese character, two models are presented for representing the structure for handwritten Chinese character recognition: the substroke center model and the stroke relation matrix model, as the classifying criteria and methods based on models. According to the proposed models, a twostage Chinese character recognition system is developed, that applies the substroke center model in the firststage and stroke relation matrix model in the secondstage. Experimental results show the feasibility and effectiveness of the proposed models.

关 键 词：手写体汉字识别 结构模型 笔段中心点 笔划关系矩阵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]