基于优先像素与卡尔曼滤波追踪的车道线检测

发布时间：2018-01-02 20:28

本文关键词：基于优先像素与卡尔曼滤波追踪的车道线检测　出处：《汽车工程》2016年02期 　论文类型：期刊论文

【摘要】：针对车道线检测不能满足实时性与鲁棒性要求的问题,提出了一种新的车道线检测方法。基于R,G和B三原色在灰度图像中所占比例的多样性、车道线与道路的亮度差,将黄色、白色像素作为优先像素处理。首先通过图像的形态学礼帽算法去除大量噪声,再经最大类间方差法(OTSU)将图像二值化,最终通过轮廓的筛选标注车道线,后续视频帧采用卡尔曼滤波追踪处理,确定新的感兴趣区域。本算法大大减少数据计算量,提高了处理效率,同时正确提取感兴趣区域,提高了算法的鲁棒性,降低车道线检测的误检率。
[Abstract]:Aiming at the problem that lane detection can not meet the requirements of real-time and robustness, a new lane line detection method is proposed, which is based on the diversity of RG and B primary colors in grayscale images. The brightness of lane line and road is poor, yellow and white pixels are treated as priority pixels. Firstly, a large amount of noise is removed by morphological hat algorithm. Then the image is binarized by the method of maximum Inter-class Variance (OTSUA). Finally, the lane line is marked through the screening of the contour, and the subsequent video frames are processed by Kalman filter tracking. The algorithm greatly reduces the amount of data computation and improves the processing efficiency. At the same time, it can extract the region of interest correctly and improve the robustness of the algorithm and reduce the false detection rate of lane line detection.
【作者单位】：湖南大学汽车车身先进设计制造国家重点实验室;
【基金】：湖南省自然科学基金(14JJ3055) 国家863计划项目(2012AA111802)资助
【分类号】：U463.6;TP391.41
【正文快照】： 前言随着汽车工业的发展,高速公路交通事故也呈现出增长的趋势,调查报告显示,50%的交通事故是由车道偏离引发的。轨道偏离预警系统能在驾驶员由于疲劳驾驶而偏出车道的时候提前报警,提醒驾驶员回到正常行驶的车道内,促进安全驾驶。目前,基于视觉的轨道偏离预警系统[1-3]占主导

【相似文献】