基于多群体遗传算法的汉字识别系统设计  

Design of a Chinese character recognition system based on multi-population genetic algorithm

作　　者：王建平[1] 陈正伟[1] 栾庆磊[1] 金铁江[1] 

机构地区：[1]合肥工业大学电气与自动化工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2010年第1期42-46,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家试验教学示范中心资助项目(411101)

摘　　要：文章设计了一种基于粗糙集理论的手写体汉字识别决策系统,依据并行遗传算法在处理大型数据库上的特有优势,将信息熵理论作为粗糙集不确定性的度量,给出一种约简冗余条件属性求解决策信息系统约简集的多群体并行遗传算法;并提出了一种基于规则置信度的手写体汉字识别规则融合算法,提高识别准确率,对SCUT_IRAC手写体汉字库中手写体汉字进行识别实验,结果证明,该系统具有较强的手写体汉字识别能力。A rough set-based recognition decision system is constructed.The entropy theory is used to measure the uncertainties of rough sets.By taking the advantages of the parallel genetic algorithm in dealing with huge data sets,a multi-population genetic algorithm is presented to reduce the overlapped attribute of the decision information system to get the reduced set.A kind of rule fusion method for Chinese character recognition based on confidence is designed to enhance the rate of accuracy.Handwritten Chinese characters from SCUT-IRAC HCCLIB are tested and the results confirm that the proposed method has the ability of recognizing handwritten Chinese characters.

关 键 词：手写体汉字识别 属性约简 遗传算法 粗糙集 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]