基于图像验证的ETC干扰问题解决方法  

A solution to interference problems in ETC systems based on image verification

作　　者：王亮[1] 鲁华祥[1] 景为平[2] 陈天翔[3] 

机构地区：[1]中国科学院半导体研究所,北京100083 [2]南通大学江苏省集成电路设计重点实验室,江苏南通226019 [3]厦门理工学院电子与电气工程系,福建厦门361005

出　　处：《智能系统学报》2012年第1期50-55,共6页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金资助项目(61076014);江苏省高校自然科学基金资助项目(10KJA510042);厦门市重大科技计划项目(3502Z20111008)

摘　　要：电子不停车收费系统(ETC)的干扰问题的原因在于现有的技术验证系统通过微波通信没有获得正确的车辆信息,针对这个问题提出一种基于字符凹凸特征的快速牌照字符验证方法.该方法以系统获得的牌照信息为已知信息,选择当前车辆牌照字符图像的凹凸特征验证其是否来源于当前待交易车辆,无需对牌照图像进行复杂的预处理,也不是直接的牌照字符识别,从而满足了ETC系统的实时性要求.实验结果表明,采用该方法以及车辆颜色和车型验证,能可靠地检查出非法车辆,切实解决ETC系统的干扰问题.The cause for interference problems in electronic toll collection（ETC） systems is a lack of a means to verify the vehicle information collected through microwave communication.A fast license plate verification method based on the character image＇s convex-and-concave features was proposed in this paper.This method uses the vehicle information collected by an ETC system as known information and chooses the character image＇s convex-and-concave features to verify whether the vehicle information originates with the car in question.Because this method needs no complicated preprocessing and does not involve direct character recognition,it complies with the ETC system＇s real-time requirements.Experimental results show that this method,supplemented by color and size verification,can reliably detect illegal vehicles and solve interference problems in ETC systems.

关 键 词：ETC 牌照验证 凹凸特征 跟车干扰 邻道干扰 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]