基于matlab的车牌定位系统的研究与设计

基于Matlab的车牌定位系统的研究与设计

本科毕业设计

基于Matlab的车牌定位系统的研究与设计

基于Matlab的车牌定位系统的研究与设计

摘 要

随着我国城市急速发展，交通管理成了一个非常热门的话题，由于交通管理需求日益提高，对交通管理和控制提出了更高的要求。在智能化交通管理系统中，车牌识别技术(License Plate Recognition，LPR) 是实现车辆识别和监控的一种重要技术，在城市车辆管理上发挥着十分重要的作用。车牌识别技术广泛应用于道路违规车辆监控、停车场管理、车辆安全管理等领域。车牌识别软件系统一般分为三部分：车牌定位、字符分割和字符识别。本文对车牌定位算法和字符分割方法作出研究和阐述，针对车牌的特点提出了一套简单的车牌定位和字符分割算法。

车牌定位部分采用的是数学形态学的方法，通过以下几个步骤实现：先利用radon函数对图像进行倾斜检测和校正，再通过灰度化和边缘检测处理之后就可以进行数学形态学处理，最后就是车牌区域的提取。字符分割部分采用的是投影法，分析垂直投影的像素点直方图，利用每个字符的波谷、波峰和车牌的定制标准来划分每一个字符。利用MATLAB强大的图像处理能力，通过编写程序，对车辆图像进行统计分析，实现了这两个部分的功能。

车牌定位算法的关键是选用合适的结构元素，对车牌图像进行数学形态学开、闭运算来去除干扰因素、消除噪音和无用信息，简单直观，受环境干扰少。经过实验证明，本文提出车牌定位算法和字符分割方法均能达到良好的效果。 关键词：车牌定位 字符分割 数学形态学 投影法

基于Matlab的车牌定位系统的研究与设计

Research On License Plate Location System

Zhou Rujie

(College of Engineering, South China Agricultural University, Guangzhou 510642,China) Abstract: With the rapid development of our country city, traffic management has become a very hot topic, due to traffic management and increasing demand, put forward higher request to the traffic management and control. The traffic management system in intelligent, license plate recognition technology (License Plate Recognition, LPR) is an important technology of vehicle identification and monitoring, play a very important role in the city vehicle management. License plate recognition technology is widely used in road illegal vehicle monitoring, parking management, vehicle safety management. License plate recognition software system is divided into three parts: license plate location, character segmentation and character recognition. This paper makes research and discusses the algorithm of license plate location and character segmentation method, according to the characteristic of license plate of a vehicle license plate location and character segmentation of simple algorithm is proposed.

License plate location part uses the method of mathematical morphology, is realized by the following steps: first use the radon function to skew detection and correction of the image, and mathematical morphology processing by gray and edge detection, then can be carried out, finally is the extraction of license plate region. Character segmentation is made using projection method, analysis of pixel vertical projection histogram, with each character trough, wave and plate custom standards to divide each character. Using image processing capabilities of the MATLAB, through the preparation process, statistical analysis of vehicle image, to achieve the functions of two parts.

The key of this method is the selection of appropriate structure elements, the license plate image mathematical morphology open, close operation to remove the interference factors, eliminating the noise and useless information, simple and intuitive, less interference by the environment. Through the experiment, in this paper, license plate localization algorithm and character segmentation method can achieve good effect.

Key words: license plate location character segmentation mathematical morphology projection