基于分级网络的车牌字符识别算法  

Pattern recognition algorithm for license plate characterbased on multilevel network

作　　者：李孟歆[1,2] 吴成东[2] 

机构地区：[1]沈阳建筑大学信息与控制工程学院,沈阳110168 [2]东北大学信息科学与工程学院,沈阳110004

出　　处：《计算机应用研究》2009年第5期1703-1705,共3页Application Research of Computers

基　　金：国家科技部国际合作重点项目(2003DF020009);沈阳市人才基金资助项目

摘　　要：提出了一种基于分级径向基神经网络的车牌字符识别算法,采用两级径向基神经网络结构,由一级网络识别后,通过对识别结果和置信度进行分析,确定了12个二级子网;采用动态误差分割算法进行网络训练,通过大量实验研究,车牌整体识别率达到了86.58%,平均识别时间为181ms,对比性研究验证了该方法的有效性。This paper proposed a new recognition algorithm for license plate character based on multi-level network, adopted two-level network. According to recognition results from one-level network and the confidence levels, built recognition distri-bution table, and accordingly designed two-level network. As a result, formed 12 two-level sub-networks. A large amount of samples are used for system test. Overall recognition accuracy is 86.58% and recognition time is 181ms. Through contrastive research, the method nresented is proved to be effective.

关 键 词：车牌识别 径向基函数网络 二级网络 识别率 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP391.41[自动化与计算机技术—计算机科学与技术]