文档介绍：苏州大学
硕士学位论文
基于视频图像处理的高速公路车辆测速算法的实现
姓名:钟敏
申请学位级别:硕士
专业:电子与通信工程
指导教师:侯嘉
2010-10
基于视频图像处理的高速公路车辆测速算法的实现中文摘要
基于视频图像处理的高速公路车辆测速算法的实现
中文摘要
基于视频图像处理的交通信息检测算法及其系统因为其方便性、实时性好,近年
来已经成为国内外交通信息领域研究的热点。车辆测速是交通信息检测的一个重要组
成部分,本课题以实时的车辆行驶速度作为研究检测对象,为道路交通管理提供更为
有效的数据,提高交通管理的技术水平。
在本课题测速算法中,车辆的检测基于车道,同时利用霍夫变换实现了像平面的
车道分割线二维重建。另外算法还建立了地面、摄像机、成像平面坐标系,用来提
取线性拟合后的车道图像特征参数进行云台摄像机的标定,在算法中,应用了自适应
背景减除、扩展 Kalman 滤波器等方法来提取运动目标及其轮廓,从而实现了车辆的
精确定位、跟踪,以至实时速度检测。
利用本文介绍的车速测量算法,在沪宁高速公路管理系统中实现并应用,结果表
明其具有较好的鲁棒性。
关键词: 车速检测, 图像处理, 摄像机标定, 车辆跟踪
作者:钟敏
指导教师:侯嘉
I
Processing of Expressway real-time speed Detection algorithm Based on Video Abstract
Processing of Expressway real-time speed Detection
algorithm Based on Video
Abstract
Traffic information surveillance algorithm based on video processing is ing
more and more important because of its convenient, timeliness. Vehicle speed detection is
very important part in the traffic information surveillance system. The paper makes the
running speed of vehicles as the studying and detection object. The purpose of this subject
is to provide more efficient references to increase the technology level of traffic
management.
In this paper, the vehicle detection is based on the roadway. With Hough transform,
the slopes of linear lane boundaries at the bottom of images are estimated, and the vanish
point of these lane boundaries are also distilled. Given these parameters, the camera
calibration is performed to remove the perspective effect of the ing images. And then,
the speed of vehicles can be calculated based on their positions of the image plane from the
vehicle tracker which realizes segmenting and real-time tracking of moving targets using
background subtraction method and the Kalman filters. The experimental results
demonstrate our algorithm, with great facility in camera calibration.
Using this algorithm is introduced in this paper, in the shanghai-nanjing expressway
man