期刊文献+

铁路线路障碍物判识技术及应用

The Technology of Identifying Railway Obstacles and Its Application
下载PDF
导出
摘要 西南铁路沿线环境复杂,降雨、地震时常引发岩石等物体滚落在线路之上,危及高速铁路及普速铁路的行车安全。为保障行车安全,中国铁路成都局集团在沪昆铁路等设立了异物侵限监测报警系统,然而列车临停、小动物上道、人工作业等时常引发系统报警,拦停列车或干扰现场作业。本文介绍了基于三维模式的线路障碍自动监测报警系统,该系统采用三维激光雷达对线路设定的防区进行扫描得到点云数据,线路两边同时增设视频监控设备得到图像数据,通过激光雷达点云数据与图像数据融合,对侵限物体目标实现多类别的识别,过滤报警干扰项,有效地避免了铁路线路限界内障碍物监测报警系统存在无效报警情况,极大地降低了误报对行车的干扰。 The environment along the railways in southwest China is complex. Rainfall and earthquakes often cause rocks and other objects to fall on the line,endangering the traffic safety of high-speed railways and conventional railways. In order to ensure train operation safety,China Railway Chengdu Group Co.,Ltd. has set up a foreign object intrusion monitoring and alarm system for railways such as Shanghai-Kunming Railway. However,temporary stops of trains,small animals on the track,worker’s operation,etc. often trigger system alarms,stop trains or interfere with on-site operations.This paper introduces an automatic monitoring and alarm system for railway obstacles based on a three-dimensional mode.The system uses three-dimensional laser radar to scan the defense area set for the line to obtain point cloud data,and video monitoring equipment is added on both sides of the line to obtain image data. Through the fusion of laser radar point cloud data and image data,multi-category identification of intrusion objects is realized. Filtering alarm interference items effectively avoids invalid alarms in the obstacle monitoring and alarm system within the clearance of railway lines and greatly reduces the interference of false alarms to train operations.
作者 郑亚宏 Zheng Yahong(China Railway Chengdu Group Co.,Ltd,Chengdu 610081,China)
出处 《高速铁路技术》 2022年第5期64-68,共5页 High Speed Railway Technology
关键词 铁路线路 障碍监测 激光点云 图像识别 数据融合 railway line obstacle monitoring laser point cloud image recognition data fusion
  • 相关文献

参考文献7

二级参考文献35

  • 1王剑英,齐亚斌.脉冲激光雷达测距系统虚警分析[J].红外与激光工程,2006,35(z3):257-261. 被引量:4
  • 2张宇,唐勐,陈锺贤,吴杰,孙秀冬,苏建忠,刘锋,崔玉平.直接探测激光雷达系统的建模与仿真[J].红外与激光工程,2004,33(6):572-575. 被引量:18
  • 3郑效田,杜立志,徐乐年,刘昆.矿井水文自动监测报警系统的设计和实现[J].煤矿机械,2006,27(9):144-146. 被引量:8
  • 4邬永革,机器人,1995年,17卷,增刊,564页
  • 5Chang H D,韩国 Pohang University of Science and Technology 技术报告
  • 6Liu K,SPIE
  • 7ELGAMMAL A, HARWOOD D, DAVIS L. Non-para-metric Model for Background Subtraction [C]//Proceed- ings of the Europe Conference on Computer Vision. Zur- ich, Switzerland: ECCV, 2000: 751-767.
  • 8TOYAMA K, KRUMM J, BRUMITf" B, et al. Wallflow- er: Principles and Practice of Background Maintenance [C]//Proceedings of the 17th IEEE International Confer- ence on Computer Vision. New York: IEEE Prees, 1999: 255-261.
  • 9LI L, HUANG W, GU I Y, et al. Foreground Object De- tection from Videos Containing Complex Background [C]// Proceedings of the the llth ACM International Conference on Multimedia. New York: ACM,2003: 2-10.
  • 10OLIVER N M, ROSARIO B, PENTLAND A P. A Bayes- ian Computer Vision System for Modeling Human Interac- tions [J]. IEEE Transactions on Pattern Analysis and Ma-chine Intelligence, 2000, 22(8): 831-843.

共引文献28

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部