基于独立分量分析的字符识别方法  

Character Recognition Based on Independent Component Analysis

作　　者：乌凌超[1] 莫玉龙[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200072

出　　处：《上海大学学报（自然科学版）》2003年第3期193-196,共4页Journal of Shanghai University:Natural Science Edition

基　　金：国家自然科学基金 (6 98750 0 9)资助项目

摘　　要：该文介绍了独立分量分析(ICA)的基本原理,给出了应用于字符识别的ICA算法,得到一组独立基的投影系数.在此基础上,组合欧氏距离和马氏距离实现字符分类.该方法已在汽车牌照字符识别中得到应用.This paper introduces the principle of the independent component analysis(ICA). An ICA algorithm applied to character recognition is given, which gives a group of projection coefficients on an independent basis. Character classification is realized by combining the Euclid distance and the Mahalanobis distance based on the projection coefficients. This method has been applied to character recognition for car license plate.

关 键 词：独立分量分析 字符识别 马氏距离 ICA算法 汽车牌照字符识别 

分 类 号：TP391.43[自动化与计算机技术—计算机应用技术]