基于限定区域和Hough变换的车道线的检测方法

车道线检测和预警是智能车辆视觉导航系统中的重要组成。为提高检测的实时性,提出一种基于限定区域Hough变换的车道线检测方法。首先利用车道线角度和范围的约束对车道线进行的限定区域划定,然后在区域内检测到符合条件的车线时再进一步建立动态ROI区域,使区域最小化,减少图像空间中被投票的目标点数,缩小Hough变换的投票空间。针对不同的光照情况,根据图像的平均灰度值来设计自适应预处理步骤,提高了算法抗干扰性。实验结果表明,在结构化的高速公路上,算法具有较好的实时性和抗干扰性。

Lane Detection Approach Based on Limited Region and Hough Transform

Lane detection and warning is an important ingredient of intelligent vehicle vision navigation system.In order to enhance the real-time of lane detection,a method based on Hough Transform（HT） of limited region is proposed.Firstly,deviding limited region by angle and range constraints of lane,and then,if there is eligible lane in the region,dynamic region of interest（ROI） is established,as a result region of HT is minimized and the total edge points is further reduced.With the different illumination situation,adopting corresponding selfadaptive measure in image preprocessing according to average gray,a result the noise immunity of algorithm is improved.The experiment results show that the approach has good real-time and noise immunity for the structured highway.