摘要ITS(Intelligent Transportation System)已经被证明是一种有效解决交通问题的方法,它具有处理速度快、监视范围广、安装维护便利、可获得更多种类交通参数等诸多优点。本文针对摄像头拍摄得到的交通序列图像,实现对车辆的跟踪和车流量的计数。61718
论文首先介绍了VC++和OpenCV的软件环境和图像基础知识,然后比较近年来常用的车辆跟踪算法和车辆计数算法,并分析其优缺点。利用MFC搭建简单的平台,使用背景差分法实现背景建模,运用轮廓跟踪法实现了车辆计数,从而完成对一个三车道视频的车辆检测。最后,提出了一些准确计数的改进方法以及仍然存在的问题。
毕业论文关键词 ITS OpenCV 车辆跟踪 车辆计数
毕业设计说明书(论文)外文摘要
Title Vehicle Detection Algorithms Based on Video Technique
Abstract ITS has been proven to be an effective way to solve traffic problems, it has a high processing speed, monitor a wide range of environment and convenience of installation and maintenance, get a wider variety of traffic parameters and many other advantages. In this paper, we achieve the vehicle tracking and traffic counts on the traffic captured by the camera image sequence. Firstly, we introduced VC + + and OpenCV software environment and image basics, and then compare vehicle tracking algorithm commonly used in recent years, also vehicle counting algorithm, and analyze its advantages and disadvantages. Secondly, we build a simple platform using MFC, using background subtraction to achieve background modeling, contour tracking method to achieve a vehicle count, thus completing a three-lane video vehicle detection. Finally, we proposed some methods for accurate count and put forward some problems which still exist.
Keywords ITS OpenCV Vehicle tracking Vehicle counting
目 次
1 绪论 1
1.1 智能交通系统的发展 1
1.2 车辆检测算法研究概况 2
1.3 论文研究内容与结构安排 3
2 实验平台介绍和图像基础知识 4
2.1 Visual C++ 6.0简介 4
2.2 OpenCV简介 4
2.2.1 OpenCV的基本概念 4
2.2.2 OpenCV的配置步骤 5
2.3 图像处理的基础知识 7
2.3.1 图像的预处理技术 7
2.3.2 图像的数字化 7
2.3.3 图像的二值化 9
3 视频车辆检测算法的设计 10
3.1 运动目标检测 10
3.1.1 常用的基本目标检测算法 10
3.1.2 本文所用的目标检测算法 13
3.2 运动目标跟踪 16
3.2.1 常见的目标跟踪算法 16
3.2.2 本文采用的目标跟踪算法 17
3.3 运动目标计数 19
3.3.1 常见的车辆计数算法