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

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%.