Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (12): 20-22.DOI: 10.3778/j.issn.1002-8331.2009.12.006

• 博士论坛 • Previous Articles     Next Articles

Image recognition method about least curvature radius of rail

MA Hai-min,ZHAO Shu-xu,DANG Jian-wu   

  1. School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China
  • Received:2008-11-26 Revised:2009-01-22 Online:2009-04-21 Published:2009-04-21
  • Contact: MA Hai-min

铁轨最小曲率半径的图像识别方法

马海民,赵庶旭,党建武   

  1. 兰州交通大学 电子与信息工程学院,兰州 730070
  • 通讯作者: 马海民

Abstract: When railway train running on sections of small radius,its speeding will make great hidden trouble for security.Through analyzing the onboard video image,this paper constructs detection templates of rail,uses region growing,curve fitting,and other means to achieve calculation of the rail curvature.Automatically detect the smallest radius of curvature in front of the train.The driver acquires early warning,and ensures safe operation of trains.

Key words: rail, curvature radius, region growing, curve fitting

摘要: 铁路列车运行在小半径路段时,超速行驶会对安全产生极大隐患。通过对车载视频图像的研究,构造铁轨检测模板,利用区域生长、曲线拟合等手段实现了对铁轨曲率的计算,可以自动检测列车前方铁轨最小曲率半径,给驾驶员以预警功能,保证列车安全运行。

关键词: 铁轨, 曲率半径, 区域生长, 曲线拟合