›› 2013, Vol. 26 ›› Issue (2): 71-.

• 论文 • 上一篇    下一篇

车辆压线检测方法

王建华,徐贵力,糜长军,赵敏,田裕鹏,王彪,徐培智   

  1. (1.江苏省交通科学研究院 股份交科院,江苏 南京 211112;2.南京航空航天大学 自动化学院,江苏 南京 210016)
  • 出版日期:2013-02-15 发布日期:2013-06-20
  • 作者简介:王建华(1981—),男,助理工程师。研究方向:计算机视觉。E-mail:wangjianhua@jstri.com。徐贵力(1972—),男,教授,博士生导师。研究方向:计算机视觉,光电检测。

Research on Pressing Line of Vehicles

WANG Jianhua,XU Guili,MI Changjun,ZHAO Min,TIAN Yupeng,WANG Biao,XU Peizhi   

  1. (1.Academy of Shares Transportation,Jiangsu Transportation Research Institute,Nanjing 211112,China;2.College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
  • Online:2013-02-15 Published:2013-06-20

摘要:

针对车辆压线的违章行为,提出一种基于计算机视觉的车辆压线检测方法。首先利用Hough变换对交通视频图像中的黄线实现自动提取;然后用背景差法从图像中提取前景车辆,并用大津法将车辆分割出来,从而获得车辆位置;最后根据黄线区域和车辆区域是否有重叠来判断车辆是否压线。实验结果表明,该方法能满足准确高效的要求。

关键词: 图像处理, 直线提取, 车辆检测

Abstract:

Aiming at detecting the violation behavior of vehicle's line pressing,this paper proposes a method of vehicle's pressing line detection based on computer vision.First,the Hough transform method is applied to extract the yellow lines from traffic video images automatically;second,the background difference method is used to extract vehicles from images,and the otsu method segments the vehicles out,thus getting vehicle's position;Finally,the overlap of the yellow line area and vehicle area can determine the vehicle's line pressing.Experimental results demonstrate that this method can meet the requirements of accuracy and efficiency.

Key words: image processing;line extraction;vehicle detection

中图分类号: 

  • TP391.41