毕业设计（论文）-C++6.0车牌识别系统设计.doc

摘 要 随着信息技术和智能技术的发展，交通管理系统的信息化、智能化已成为发展的趋势。车牌识别系统(LPR)是智能交通系统的核心组成部分，在现代交通收费管理系统中发挥着举足轻重的作用，而车牌定位又是车牌识别系统中的重要环节。近年来，对车牌识别系统中关键技术的研究已经成为智能交通领域的一个热点问题。本文对车牌识别系统中的车牌定位算法进行了比较深入、全面的分析与论述。 基于编程软件Microsoft Visual C++6.0平台，利用数字图像处理对车牌定位算法进行了实现，并得到了预期的效果。主要步骤包括以下几个方面： 1.采集到的彩色图像转换为灰度图像后，采用直方图均衡化方法对车牌图像进行图像增强。 2.使用Sobel算子对图像进行边缘处理，并进一步加深，为车牌定位做好准备工作。 3.针对车牌定位问题，采用模板匹配算法确定车牌区域，即用一个与车牌部分相似的模板图像与原图像进行比较，确定原图中有与该模板相同或相似的区域就是车牌所在的区域，进而定位出车牌。 实验结果表明，对图像进行一系列的处理后进行模板匹配，定位出了车牌所在的区域，达到了预期的目标。 关键词：车牌定位；Microsoft Visual C++6.0；数字图像处理；图像转换；模板匹配  ABSTRACT With the information technology and intelligent technology development, the traffic management system and intelligence information has become the development trend. License plate recognition system (LPR) is the core of the intelligent transportation system components, in modern transportation charges administration system plays an important role in and license plate localization is an important link in the license plate recognition system. In recent years, the system of vehicle plate recognition of the key technology research in the field of intelligent transportation has become a hot issue. In this paper, the license plate identification system of license plate localization algorithm is more thorough, comprehensive analysis and discussed. Based on the software Microsoft Visual C++ 6.0 platform, using digital image processing to license plate localization algorithm was realized, and to get the desired results. Main steps include the following aspects: The collected color images into the gray image, the method of histogram equalization license plate image of the image enhancement. Use Sobel operator to do image edge processing, and further deepen, license plate location ready for work. According to license plate location problem, a template matching algorithms are used to determine the license plate area, which is a part of the license plate with similar template image comparison with the original image, sure in