基于概率投票法的道路消失点检测算法研究  

Research on Detection Algorithm of Road Vanishing Point Based on Probabilistic Voting Method

作　　者：尹金楷[1] YIN Jin-kai(Automobile College of Sanmenxia Polytechnic,Sanmenxia 472000,China)

机构地区：[1]三门峡职业技术学院,三门峡472000

出　　处：《内燃机与配件》2021年第7期170-171,共2页Internal Combustion Engine & Parts

基　　金：课题名称:基于教师教学能力大赛的高职院校教学改革与研究--以新能源汽车技术专业为例,编号:SZY-2020-012

摘　　要：本文提出了一种稳定的基于消失点估计的车道检测方法。估计消失点有助于检测车道,因为在投影的二维图像中,平行线会聚集在消失点上。然而,在背景复杂的图像中,正确估计消失点并不容易。因此,提出了一种鲁棒的消失点估计方法,该方法使用基于从输入图像提取的线段的交点的概率投票过程。实验结果表明,该方法能有效地估计消失点,并在不同环境下检测车道。The paper proposes a stable lane detection method based on vanishing point estimation.Estimating the vanishing point helps to detect the lane,because in the projected two-dimensional image,parallel lines will gather on the vanishing point.However,in an image with a complex background,it is not easy to correctly estimate the vanishing point.Therefore,a robust vanishing point estimation method is proposed,which uses a probabilistic voting process based on the intersection of line segments extracted from the input image.Experimental results show that this method can effectively estimate the vanishing point and detect lanes in different environments.

关 键 词：消失点估计 车道检测 概率投票 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]