复杂道路环境下车道线快速提取方法  

Fast Extraction Method of Lane Lines in Complex Road Environment

作　　者：储开斌[1] 郭俊俊 朱栋 CHU Kaibin;GUO Junjun;ZHU Dong(School of Information Science and Engineering,Changzhou University,Changzhou 213164,Jiangsu,China)

机构地区：[1]常州大学信息科学与工程学院,江苏常州213164

出　　处：《实验室研究与探索》2020年第7期11-15,共5页Research and Exploration In Laboratory

基　　金：国家自然科学基金项目(61772090);江苏省重大成果转化项目(BA2017012);江苏省高等学校自然科学研究项目资助(19KJB510017);常州大学项目资助(ZMF18020066)。

摘　　要：针对无人驾驶车辆在行驶过程中受到地面污损、障碍物、光照等环境影响,导致车道线情况复杂问题,提出一种快速的车道线识别方法。首先,对采集到的图像进行预处理,设置自适应感兴趣区域,并进行灰度化、双边滤波和边缘提取。其次针对边缘噪声复杂的情况,提出了一种新的边缘噪声消除方案,将边界图像划分为若干个子图像并通过计算每个子图像的边缘方向来实现噪声边缘的消除,再对被切断的车道边缘块进行补偿。最后,通过直线拟合算法拟合出车道线。实验结果表明,所提方法可以有效地消除边缘噪声,且在各种天气和路况下具有较快的处理速度和较高的精度。The unmanned vehicles are easily affected by environmental factors such as the ground pollution,obstacles,illumination and etc.during the driving process.Therefore,a fast lane line identification method is proposed in this paper.Firstly,the acquired image is preprocessed by setting the adaptive region of interest and performing grayscale classification,bilateral filtering,and edge extraction.A new edge noise elimination scheme is proposed to deal with the condition of complex edge noise.The boundary image is divided into several sub-images on which the noise edge is eliminated by separately calculating the edge direction,and then compensating the divided lane edge blocks.Finally,the lane line is connected by a simple straight-line fitting algorithm.The experimental results indicate that the proposed method can effectively eliminate edge noise with fast processing speed and high accuracy under various weather and road conditions.

关 键 词：感兴趣区域 车道线 边缘检测 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]