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

基于粗糙集的车牌字符识别方法
引用本文:高伟,刘喜平.基于粗糙集的车牌字符识别方法[J].山西大学学报(自然科学版),2005,28(3):253-256.
作者姓名:高伟  刘喜平
作者单位:1. 山西大学,数学科学学院,山西,太原,030006
2. 山西省自动化研究所,山西,太原,030012
基金项目:山西省科技项目(编号:021092)
摘    要:提出了一种基于粗糙集理论的车牌字符识别的方法,通过粗糙集的属性约简,有效地压缩了图像的特征数目,提高了运行效率,并且采用基于影响因子的图像判别算法,有效地提高了识别的准确率.以在高速公路收费站实地拍摄的车牌图像为样本,经过车牌的定位、分割,以及字符的分割,选取其中的300幅字符图像作为训练集,100幅字符图像作为测试集,实验结果表明:将训练集图像作为输入,正确识别率为100%;将测试集作为输入,正确识别率为86%。

关 键 词:粗糙集  字符识别  车牌识别
文章编号:0253-2395(2005)03-0253-04
收稿时间:2004-05-20
修稿时间:2004年5月20日

A Method of Recognizing Characters in Vehicle License-Piates Based on the Theory of Rough Sets
GAO Wei,Liu Xiping.A Method of Recognizing Characters in Vehicle License-Piates Based on the Theory of Rough Sets[J].Journal of Shanxi University (Natural Science Edition),2005,28(3):253-256.
Authors:GAO Wei  Liu Xiping
Abstract:A method of recognizing characters in vehicle license-plates based on the theory of rough sets was introduced.By the attribution reduction of the rough sets theory,the number of information about character-images and the task's running time could be significant reduced.300 images were chosen as training sets and 100 images as testing sets.According as the method the results show that the accuracy is as high as 100 % when the training sets were regarded as inputs;the accuracy was as high as 86 % using testing sets as inputs.
Keywords:rough sets  vehicle character recognition  license plate 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