车牌识别系统的设计与实现毕业设计论文

  • 格式:doc
  • 大小:2.07 MB
  • 文档页数:49

下载文档原格式

  / 49
  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

车牌识别系统的设计与实现

摘要

车牌识别系统是智能交通系统的重要组成部分,有着广泛的应用,同时也是计算机视觉、图像处理和模式识别等交叉学科的研究热点。车牌识别系统广泛应用于高速公路自动收费和超速监管系统、公路流量监控系统、停车场收费管理系统、安防系统以及小区物业管理系统等等。

本文在VC++6.0平台下,结合OpenCV视觉库,开发了车牌识别系统。车牌识别系统分为五大组成部分:图像预处理、车牌定位、车牌倾斜校正、字符分割和字符识别。

本系统能够对输入的图像进行图像预处理,包括图像灰度化、图像滤波、边缘检测和二值化等;其次,通过使用数学形态学技术和车牌的高宽比固有特征从预处理后的图像中定位出车牌位置,并把车牌提取出来;再次,使用Hough 变换方法测出倾斜角,对车牌进行有效的旋转校正;通过水平投影去除车牌的上下边框,使用垂直投影法把字符一个个分割出来;最后,通过模板匹配法识别车牌字符并输出。实验结果表明,车牌定位的准确率为87.9%,车牌字符的正确识别率为72.6%。

关键词:车牌定位,数学形态学,Hough 变换,倾斜校正,投影

Design and Implementation of License Plate Recognition System

Abstract

License plate recognition system is an important part of intelligent transportation system and it has a wide range of applications. It is a hot spot of research in computer vision, image processing and pattern recognition. License plate recognition system is widely applied in the highway automatic charging and speed regulation system, highway traffic monitoring system, parking fee management system, security system and the residential property management system, etc.

This system is developed in VC++6.0 platform and computer vision library OpenCV. License plate recognition system designed is mainly divided into five parts: image preprocessing, license plate orientation, license plate tilt correction, character segmentation and character recognition.

This system can perform the image pretreatment to the input image, including the image gray level, the image filtering, edge detection and binarization and so on. Secondly, by using the technology of mathematical morphology and the plate inherent characteristics of aspect ratio to position license plate location in the image after pretreatment, and extract the license plate. Thirdly, Hough Transform method is used to measure the angle and effective rotation correction the license

plate. Fourthly, by using horizontal projection to remove the up and down borders of the license plate, the characters are divided one by one using the vertical projection method. Finally, using the template matching method to recognize the characters, and output it. Experiments show that the accuracy of license plate location is 87.9%, the correct rate of license plate character recognition is 72.6%.

Keywords:license plate localization, mathematics morphology, Hough transform, tilt correction, projection

目录

1引言 (1)

1.1课题研究目的与意义 (1)

1.2国内外研究现状 (3)

1.2.1国内研究现状 (3)

1.2.2国外研究现状 (4)

1.3目前存在的问题 (5)

2系统开发编程语言 (7)

2.1 Visual C++介绍 (7)

2.2 OpenCV介绍 (8)

3系统的总体设计.................................................................. 错误!未定义书签。

3.1需求分析....................................................................... 错误!未定义书签。

3.2软件系统组成............................................................... 错误!未定义书签。

3.3系统的层次结构........................................................... 错误!未定义书签。

3.4系统的软硬件环境....................................................... 错误!未定义书签。4系统的详细设计.................................................................. 错误!未定义书签。

4.1图像预处理................................................................... 错误!未定义书签。

4.1.1功能概述.............................................................. 错误!未定义书签。

4.1.2问题分析.............................................................. 错误!未定义书签。

4.1.3预处理方法及实现.............................................. 错误!未定义书签。

4.2车牌定位....................................................................... 错误!未定义书签。