基于一维离散周期小波变换的实时车牌识别方法  

Real-time vehicle license plate recognition based on 1-D discrete periodic wavelet transform

作　　者：曾文艳[1] 王亚刚[1] 邵惠鹤[1] 

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《信息技术》2014年第2期65-68,72,共5页Information Technology

基　　金：国家自然科学基金项目(61074087);上海市教育委员会科研创新项目(12ZZ144)

摘　　要：利用扫描窗口和一维离散周期小波变换以及BP人工神经网络进行车牌识别。被扫描的图像通过一维离散周期小波变换来选择图像的低频系数,这样可以提高执行车牌识别的速度。文中方法是直接对车牌进行扫描,不对单个字符进行识别,并通过MATLAB编程实现。新方法是一种实时识别,车牌识别的实验结果可高达94.7%。This paper is the use of the scanning window and 1-D discrete periodic wavelet transform and back propagation artificial neural network for the license plate recognition. By scanning the image through the one-dimensional discrete periodic wavelet transform, it selects the image low-frequency coefficients. Therefore, it can improve the rapid license plate recognition of implementation. It is a directly scanning for the license plate recognition, without the individual character recognition, and it is implemented by MATLAB. This new method is a kind of real-time,the recognition rate of the experimental results can be as high as 94.7%.

关 键 词：离散周期小波变换 车牌识别 BP人工神经网络 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]