有限范围的视频车辆识别算法的研究和实现.docxVIP

（20 届） 毕业论文（设计） 有限范围的视频车辆识别算法的研究和实现 扌商要：车牌识别系统（LPR系统）是以汽车牌照为特定H标的专用识别系 统，以其在现代化交通屮所占的重要位置，已成为智能交通领域研究的重点和热 点之一。车牌识别系统涉及到数学图像处理、计算机视觉、模式识别、人工智能 等多项技术。对车牌照的定位、分割和字符识别是最为关键的技术，对整个系统 的性能起着决定性作用。 木系统主要由车牌图像采集和预处理、车牌区域的定位提取、车牌字符的分 割和识别儿部分纽成。本文对系统屮车牌定位、字符分割和识别的算法做了一个 比较全面的论述，分析了现有的理论算法，并分别提出具有实际应用意义的解决 方案。主要解决了： 一、在简单背景图像屮准确定位分割车牌区域；二、对车牌 字符的识别能力进行提高。 在图像预处理模块对图像进行灰度化、二值化和图像增强处理。 在车牌定位模块概述了四种研究方法，提岀利用边缘检测法对牌照区域 进行定位。 在车牌分割模块，在进行了车牌的倾斜校正和确定字符上下边界后，利 用滑动窗口探测分割法实现字符分割。 在车牌识别模块，在字符特征提取的基础上提出了 BP神经网络识别方 法。通过输入特征，训练样本的学习和实验，在识别上去的较好的效果。 在本文屮，对车牌识别系统进行一些研究工作，在实际屮的应用效果还不是 很令人满意，有待进一步的研究和完善。 关键词：图像处理车牌定位字符分割车牌识别BP神经网络 The Limited range of video vehicle detection algorithm andimplementation Abstract: The License Plate Recognition System(LPRS) is the visual system whose goal is the License Plate, with its important position in occupied by the modern traffic. LPRS becomes one of the focus research in the intelligent transportation. LPRS involves techniques such as digital image processing, computer seeing, pattern recognition and artificial intelligence etc. Research on license positioning, segmentation and character recognition is the most critical technology? And it plays a crucial role in the performance of the whole system. This system is mainly made up of the vehicle license picture gathering and the pretreatment, the vehicle license region localization and extraction, the vehicle license character segmentation and the recognition. This article has made a quite comprehensive discussion about the algorithm of the vehicle license region localization, the vehicle license character segmentation and the recognition. This paper researches and analyzes the existing theories and algorithms, and puts forward a practical solution scheme of application significance on every module. Mainly solved the following questions: Firstly, accurately locating and dividing vehicle license region in the ordinary background image; Secondly, enhancing the capability of vehicle license character recogniti