多环境高性能车牌识别应用系统-电子与通信工程专业论文.docxVIP

万方数据 万方数据 摘 要 随着智能交通系统的发展，与图像智能分析技术相结合的嵌入式技术是目前 研究的热点之一。本文从工程应用角度出发，详细阐述了移动和固定两种不同应 用场景的嵌入式车牌识别系统。第一部分工作实现了基于 PC 车牌识别测试系统， 对车牌识别系统中的车牌定位、字符分割和车牌字符识别算法进行了深入研究， 第二部分工作针对移动应用场景实现了基于智能手机平台（Windows Phone 及 Android 平台）的车牌识别软件，第三部分工作针对固定应用场景在 EagleEye3 智 能相机平台上实现具有车牌识别功能的子系统。主要内容如下： 1.详细阐述智能手机和智能相机的嵌入式平台及车牌识别技术的研究现状以 及面临的问题和困难。 2.实现基于 PC 车牌识别测试系统，并分别对基于颜色与纹理相结合的快速车 牌定位算法、基于垂直投影与车牌固有特征相结合的车牌字符快速分割算法及基 于 Gabor 特征提取的车牌字符识别算法进行深入研究。 3.详细阐述了基于智能手机车牌识别系统的软硬件实现方法。采用智能相机手 机作为开发平台实现车牌识别、车主信息查询、号牌识别结果信息发送等功能。 4.详细阐述了将车牌识别算法移植到智能相机平台过程中所用到的关键技术， 实现基于 EagleEye3 智能相机平台的车牌识别功能子系统，并对 Windows Phone、 Android 及嵌入式 EagleEye3 等三种不同嵌入式平台的开发进行比较。 关键词：智能手机 智能相机 车牌定位 字符分割 字符识别 Abstract With the development of intelligent transportation system, image analysis and embedded technology are of great concern in recent research. Starts from engineering applications, the thesis propose the development of license plate recognition based on embedded platforms for fixed and mobile application scenes. First of all, based on intensive study to the license plate recognition system of license plate location, character segmentation and character recognition, a complete license plate recognition test system based on PC is proposed. Then, two mobile terminal-based (Windows OS and Android OS) license plate recognition systems for mobile application scenes are proposed.Finally, the license plate recongnition system based on Eagleeye3 embedded smart camera is studied for fixed application scenes. Current research status in the embedded development of mobile phone and smart camera and in license plate recognition technology, along with the problems and difficulties faced, were exposited in detail. A set of license plate recognition test system was proposed based on PC. This paper has done an in depth study on an effective license plate location algorithm based on color and texture analysis, a characters segmentation algorithm based on the intrinsic characteristics of license plate and projection and a character recogn