首页 | 本学科首页   官方微博 | 高级检索  
     

基于Hough变换的车道检测改进算法研究
引用本文:李明,黄华,夏建刚. 基于Hough变换的车道检测改进算法研究[J]. 计算机工程与设计, 2012, 33(4): 1638-1642
作者姓名:李明  黄华  夏建刚
作者单位:四川大学电气信息学院,四川成都,610065
摘    要:车道检测是车道侧偏检测的前提,是车辆辅助驾驶系统的重要组成部分,针对这一特点,提出了利用Hough变换来实现对车道检测的直道检测,而对于弯道部分的检测,则是该算法的一个创新之处.通过统计较小区域的目标像素数目,设置感兴趣区域(ROI)来判断车道的弯曲方向,并在该方向上进行弯道像素的搜索,采用多段折线的方法对弯道部分进行重建.实验结果表明该算法具有运算速度快、准确性高、鲁棒性较好的特点.

关 键 词:车道检测  Hough变换  感兴趣区域  多段折线  偏离点

Research of robust lane detection algorithm based on Hough transform
LIMing , HUANG Hua , XIA Jian-gang. Research of robust lane detection algorithm based on Hough transform[J]. Computer Engineering and Design, 2012, 33(4): 1638-1642
Authors:LIMing    HUANG Hua    XIA Jian-gang
Affiliation:(School of Electrical Engineering and Information,Sichuan University,Chengdu 610065,Sichuan)
Abstract:Aiming at lane detection,which is the premise of lane departure,plays an important roll in intelligent vehicle navigation.Hough transform is proposed to achieve the straight lane detection.While,for the corner part of the lane detection,that is one of the paper’s innovative points.Firstly,by statistic the number of pixels of small area,the interested region(ROI) is set to determine the bending direction of the lane,and then the pixels of curve lane in the direction of corners pixels is searched,finally the corners parts of the lane rebuild by using multiline methods.Experimental results show that this method has a fast calculation speed,accuracy,high robustness.
Keywords:lane detection  Hough transform  ROI  multi-line  deviation from point
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号