有代码的论文.doc

车牌提取系统的设计与实现 摘 要 汽车牌照自动识别系统是制约道路交通智能化的重要因素,包括车牌定位、字符分割和字符识别三个主要部分。本文首先确定车辆牌照在原始图像中的水平位置和垂直位置,从而定位车辆牌照,然后采用局部投影进行字符分割。在字符识别部分,提出了在无特征提取情况下基于支持向量机的车牌字符识别方法。实验结果表明,本文提出的方法具有良好的识别性能。 Car license plates recognition system is restricted by the road traffic intelligent important factors, including license plate localization, character segmentation and character recognition three main parts. This paper first determine the vehicle plate in the original image level and vertical position, thereby positioning vehicle licensing, then use local projection for character segment. In character recognition, and puts forward the part without feature extraction cases based on support vector machine plate character recognition method. Experimental results show that the proposed method has good recognition performance. With the growing popularity of highway in China, road traffic undertakings have developed rapidly, so artificially management mode is not filled with practical needs, microelectronics, communications and computer technology in traffic areas of application greatly improve the traffic management efficiency. Car licence automatic identification technology has been widely used. This article introduces the license plate extraction systems use the design principle and key technology, detailed design steps and system implementation. Keywords: license plate extracted; Image recognition; Design and implementation 目 录 摘 要 1 ABSTRACT 2 目 录 3 1 相关技术介绍 3 1.1 MFC(微软基本类库) 3 1.2 GDI(图形设备接口) 4 1.3 MFC开发软件 5 2 设计原理 5 2.1 系统流程 5 2.2 总体设计步骤 6 3 详细设计 7 3.1 预处理及边缘提取 7 3.1.1 图象的采集与转换 7 3.1.2 边缘提取 8 3.2.1 牌照区域的定位 10 3.2.2 牌照区域的分割 11 3.2.3车牌进一步处理 11 3.3 字符的分割与归一化 12 3.3.1 字符分割 13 3.3.2 字符归一化 13 3.4 字符的识别 13 4 关键代码 15 4.1 图像灰度化 15 4.2 灰度均衡化 18 4.3 边缘检测 20 4.4 预处理 23 4.5定位车牌 25 4.6 分割车牌 32 5 设计结果及分析 33 参考文献 34 1 相关技术介绍 1.1 MFC(微软基本类库) MFC是Win?API与C++的结合,API,即微软提供的Windows OS(操作系统)下应用程序的编程语言接口,是一种软件编程的规范,但不是一种程序开发语言本身,可以允许用户使用各种各样的第三方(如我是一方,微软是一方,Borland就是第三方)的编程语言来进行对Windows?OS