首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于轮廓结构和统计特征的字符识别研究
引用本文:贾婧,葛万成,陈康力.基于轮廓结构和统计特征的字符识别研究[J].沈阳师范大学学报(自然科学版),2006,24(1):43-46.
作者姓名:贾婧  葛万成  陈康力
作者单位:同济大学,电子与信息工程学院,上海,200092
基金项目:本文受到德国罗德与施瓦茨公司资助
摘    要:车牌识别系统是智能交通不可缺少的一部分,在车牌识别系统中,字符的特征提取和字符识别是这一系统的关键技术.文章利用字符的轮廓结构特征和统计特征对字符进行识别,根据字符外部轮廓的上、下、左、右4个方向的特点和一些统计特征,如字符最大宽度,垂直方向笔画数等作为识别特征,形成判别树,再利用判别树对汽车牌照中的数字和字母进行识别.此方法识别的准确率比较高,并提高了识别速度,还适用于其他不同字体的数字和字母识别.

关 键 词:结构特征  统计特征  字符识别
文章编号:1008-374X(2006)01-0043-04
收稿时间:2004-11-10
修稿时间:2004年11月10

Character Recognition Based on Structural and Statistical Features
JIA Jing,GE Wan-cheng,CHEN Kang-li.Character Recognition Based on Structural and Statistical Features[J].Journal of Shenyang Normal University: Nat Sci Ed,2006,24(1):43-46.
Authors:JIA Jing  GE Wan-cheng  CHEN Kang-li
Institution:College of Electronic and Information Engineering, Tongji University, Shanghai 200092, China
Abstract:The car plate recognition system is an indispensable part of intelligent traffic system.In car plate recognition system,one of the most difficult problems is the feature and recognition of car plate character.This paper introduces a character recognition algorithm based on structural and statistical features of character's figures.According to the features on all sides of top,bottom,left and right,as well as some statistic features,such as the width of characters,the count of strokes across the vertical direction and so on,we have designed a tree structure to recognize the numbers and leters from the license plate.This algorithm can recognize the numbers and letters successfully in a shorter time.In addition,it can be used to recognize other kinds of characters.
Keywords:structural features  statistical features  character recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号