基于视频的车辆测速技术研究计算机应用技术专业论文.docxVIP

哈尔滨工业大学工学硕士学位论文 哈尔滨工业大学工学硕士学位论文 II II Abstract With the rapid development of economic, the amount of vehicle is becoming larger and larger. How to make improvement on the status of the situation becomes the focus that the international governments and the departments pay attention to. And then, the technology of Intelligent Transportation System (ITS) is developing quickly. The system based on video is becoming the major direction of the traffic detection. This thesis is based on the background of vision-based vehicle detection and vehicle speed detection. In the paper, we focus our research on vision-based day and night distinction, image process, vehicle detection and track. Then we propose a accurate, stable vision-based vehicle speed detection method. There are different arithmetic bettwen day and night, so we propose a criterion based on time testing and criterion of gray scale. The experiment indicates that the technology has a fast operation speed, a excellent adaptable ability and a satisfactory accuracy. By day, background without any vehicle can be reconstructed from video sequences in day arithmetic and object can be detected when we subtract background from current image. At the same time, we reinforce this characteristic through the technique of frame subtraction between current frame and the last one. At night, we use a method of using morphological characteristics to distill headlights blocks. Matching the blocks and location the headlights, we will get the position of the vehicle. In the paper, we compared new arithmetic of vehicle tracking. Then we performed an object region tracking method based on its characters such as area, location and velocity. The method can satisfy the need of speed detection and efficiency at the same time. Keywords ITS, Day and night division, Motion object detection, Object track PAGE PAGE IV 目 录 摘 要 I Abstract II 第 1 章 绪论 1 1.1 课题背景 1 1.2 国内外研究现状 2 1.2.1 智能交通研究现状 2 1.2.2 车速检测研究现状 2 1.3 本文主要研究内容 4 第 2 章 视频测速理论基