无人驾驶车道检测技术的研究  

Research on driverless lane line detection

作　　者：胡晓霞[1] 李雪梅 白燕燕[1] 李瑞 HU Xiao-xia;LI Xue-mei;BAI Yan-yan;LI Rui(North Institute of Information Engineering,Xi’an Technological University,Xi’an 710025,China)

机构地区：[1]西安工业大学北方信息工程学院,陕西西安710025

出　　处：《电子设计工程》2020年第18期118-121,126,共5页Electronic Design Engineering

基　　金：国家自然科学基金(61671362);西安工业大学北方信息工程学院2018年度院长基金项目(18YJZ07)。

摘　　要：针对无人驾驶车道保持的问题,提出了一种在MATLAB平台下,基于Hough变换的无人驾驶车道线检测算法。首先,为了提高车道线检测的准确性,对采集到的道路图像进行预处理;然后,通过边缘检测提取车道线特征信息;最后,在分析现行车道线检测算法的基础上,提出了基于Hough变换的车道线检测算法的总体思路;而后讨论了该算法在实时道路检测中的可行性及检测精度,其检测精度达到82.5%,并具备较高的稳定性;实验结果表明所提出的车道线检测方法是有效的。A new driverless lane line detection algorithm based on Hough transform is proposed under the MATLAB platform to solve the problem of driverless lane keeping system.Firstly,in order to improve the accuracy of lane line detection,the collected road image is preprocessed;Secondly,the feature information of lane line is extracted by edge detection;Finally,based on the analysis of the current lane line detection algorithms,the general idea of the lane line detection algorithm based on improved Hough transform is proposed;Then,the feasibility and accuracy of the detection algorithm based on the Hough transform is applied to real-time road detection are discussed.The detection accuracy is 82.5%,and it has high stability;Experimental results show that our proposed method is effective.

关 键 词：无人驾驶 车道检测 HOUGH变换 SOBEL算子 边缘检测 

分 类 号：TN911[电子电信—通信与信息系统]