一种基于分级RBF网络的车牌字符识别方法  

Pattern recognition method for license plate character based on multilevel RBF network

作　　者：李孟歆[1] 吴成东[1] 夏兴华[2] 

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

出　　处：《计算机工程与应用》2008年第30期213-216,共4页Computer Engineering and Applications

基　　金：科技部国际合作重点项目(No.2003DF020009)。

摘　　要：提出了一种基于分级RBF神经网络的车牌字符识别方法,采用两级RBF神经网络结构,由一级网络识别后,根据识别结果和置信度,建立识别分布图,进行二级网络设计,确定了12个二级子网。RBF网络中自动确定隐层神经元数,无需实验调整。用大量样本对系统进行测试,车牌整体识别率达到了85.4%,通过对比性研究,验证了该方法的有效性和先进性。A new recognition algorithm for license plate character based on multi-level RBF network is proposed.Two-level RBF network is adopted.According to recognition results from one-level network and the confidence levels,recognition distribution table is built,and two-level network is accordingly designed.As a result, 12 two-level sub-networks are formed.A large amount of samples are used for system test.Overall recognition accuracy is 85.4%.Through contrastive research,the method presented is proved to be effective and advanced.

关 键 词：车牌识别 径向基函数(RBF)网络 二级网络 识别率 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]