摘要
从字符特点角度出发,提出采用聚类分析与神经网络方法分别对车牌中英文与数字字符、汉字字符进行识别。本文方法可以提高汉字的识别准确率,并加快车牌的识别速度,以适应高速公路收费系统即时、准确的要求。实验结果表明,数字及字母的识别准确率达97.0%,汉字的识别准确率达90.1%。
A method for license plate recognition based on clustering analysis and neural network is introduced from the Chinese,number and alphabet character.The method can improve the ratio of Chinese recognizing accuracy and increase the speed of the license plate recognition,thus satisfing the demand of immediacy and accuracy in the system of expressway toll collection.Experimental results show that the recognition accuracy of number and alphabet is 97.0% and that of Chinese character is 90.1%.
出处
《数据采集与处理》
CSCD
北大核心
2008年第2期238-242,共5页
Journal of Data Acquisition and Processing
关键词
车牌识别
神经网络
动态聚类
license plate recognition
neural network
dynamic clustering