毕业论文（设计）车牌号码识别系统的设计与实现.doc

车牌号码识别系统的设计与实现 ? 摘 要 车牌号码识别系统是利用计算机视觉系统实现的，它能够通过给定的车辆图像信息，自动识别车牌号码。该技术可应用在高速公路收费管理、停车场管理、智能化交通系统等不同场合。车牌识别系统主要包含三个大模块：字符预处理、分割和识别。 论文首先概述了该系统设计与实现的软硬件条件，接着剖析了系统设计的可行性，然后依次介绍了系统的组成模块：车牌图像预处理模块实现了图像灰度化、对比度增强，图像二值化和车牌倾斜度校正等功能；车牌字符分割模块实现了字符分割和字符预处理的功能，该模块主要是为下一步提取到精确的字符特征做准备；字符识别模块采用改进的模板匹配公式，提高了识别的效率。最后，对三个模块分别做了单元测试后把三个模块集成在一起进行了测试，分析系统操作的结果。通过测试多幅样本图像表明，系统稳定性强，达到了一定的识别率。 【关键词】?车牌图像预处理；字符分割；字符识别；模板匹配  Development and Implementation of Vehicle License Plate Recognition System Abstract License plate number recognition system is implemented by a computer vision system, it can be given by the vehicle image information automatically identify license plate numbers. The technology can be used in highway toll management, parking management, intelligent transportation systems and other occasions. License plate recognition systems typically contain license plate image preprocessing, license plate character segmentation, character recognition three modules. Firstly, analyzes the design and realization of hardware and software conditions, and then analyzes the feasibility of the system design, followed by a detailed presentation of the building blocks of the system: the license plate image preprocessing module implements the gray image, contrast enhancement, image two value and license plate tilt correction function; contain character segmentation and character segmentation character preprocessing, the module is mainly for the next character feature extraction to accurately prepare; character recognition module uses an improved template matching formula, to obtain better recognition rate, each of these modules will last for a simple system integration and testing and analysis of the system operation results. By testing multiple sample images show that the system has strong stability and achieve a certain recognition rate. [Keywords] License plate image pretreatment; Character segmentation; Character recognition; Template matching 目 录 1 引言 1 1.1 研究背景及意义 1 1.1.1 研究背景 1 1.