欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

JASE

• 汽车安全 • 上一篇    下一篇

基于机器视觉与信息共享的交叉路口交通安全预警

朱淑亮1,2，于 涛1，李 峻1   

  1. （1. 烟台大学 机电汽车工程学院，烟台264005，中国；2. 清华大学 汽车安全与节能国家重点实验室，北京100084，中国）
  • 收稿日期:2017-11-27 出版日期:2018-06-30 发布日期:2018-07-04
  • 作者简介:第一作者 / First author ： 朱淑亮（1982—），男（汉），山东，讲师。E-mail: zhuslyt@126.com。
  • 基金资助:

    汽车安全与节能国家重点实验室开放基金（KF16242）；山东省高等学校科技计划项目（J17KA039）。

Traffic safety early warning at intersection based on machine vision and information sharing

ZHU Shuliang1,2, YU Tao 1, LI Jun1   

  1. (1. School of Electromechanical and Automotive Engineering, Yantai University, Yantai 264005, China; 2. Tsinghua University, State Key Laboratory of Automotive Safety and Energy, Beijing 100084, China)
  • Received:2017-11-27 Online:2018-06-30 Published:2018-07-04

PDF

757

摘要：

       为提高道路交叉路口车辆及行人的通行安全性，提出了一种基于机器视觉与信息共享的全局域内的人车安全预警系统。该系统分为路侧设备和车载设备，通过路侧摄像机获取全路口实时视频图像，应用机器视觉技术解算车辆和行人的运动状态，通过计算碰撞时间预测监控目标继续通行的安全程度，路侧设备向危险车辆精准传递报警信息。预警系统向危险行驶车辆精准传递报警信息。对预警系统中的目标检测及运动分析进行了仿真和测试。试验表明：该预警方案在车辆检测、运动分析和安全预测方面在给定测试条件下具有可行性。

关键词: 道路交通 , 主动安全 , 交叉路口 , 驾驶辅助 , 人车路一体化（V2X）, 机器视觉 , 信息共享

Abstract:

A safety early warning system of vehicles and vulnerable road users in the global domain was proposed based on machine vision and information sharing to improve the traffic safety at road intersections. The system consisted of roadside unit and on board unit. The roadside unit cameras captured intersection real
time videos. The machine vision technology calculated the motion states of vehicles and vulnerable road users. The traffic safety degrees of the monitoring targets were judged by calculating the time to collision. The roadside unit released alarm information precisely to the dangerous vehicle. The system transmitted accurately warning information to the dangerous vehicle. The target detection and the motion analysis were simulated and tested for the system. The test results show that the early warning scheme is feasible in vehicle detection, motion analysis and safety prediction under the test conditions.

Key words: road transport, active safety , intersection, driving assistance , vehicle to X (V2X) ,  machine vision , information sharing