机电工程技术2024，Vol.53Issue(4)：111-114,4.DOI:10.3969/j.issn.1009-9492.2024.04.025
基于霍夫变换与最小二乘法的车道线检测方法
Lane Line Detection Method Based on Hough Transform and Least Squares Method
摘要
Abstract
Based on traditional image processing methods,image pre-processing,ROI segmentation,Hough transform and least square method are used to detect lane lines.Preprocessing the image,which includes binarizing the image;then edge detecting the image through Gaussian filter.In order to improve the execution efficiency and accuracy,the ROI area displayed under the 15°top view angle is used for detection,and the Hough transform is used to detect the lane line.By setting thresholds and adjusting parameters,the peak points of the lane line are obtained,and the least squares method is used to fit the straight line.The detection results are drawn in the image,including lane lines and Hough transform peak points.By using this method,experiments were conducted on five experimental groups on three routes,including highways,urban roads,and provincial roads.The accuracy of highways and provincial roads is 95%and 98%respectively,and the accuracy of urban roads is 75%.The results show that in specific environments,by using the method proposed,the lane lines can be effectively detected,it has good robustness and accuracy.关键词
车道线检测/ROI/霍夫变换/最小二乘法Key words
lane detection/ROI/Hough transform/least squares分类
交通工程引用本文复制引用
廖干洲,曾霞,李龙华..基于霍夫变换与最小二乘法的车道线检测方法[J].机电工程技术,2024,53(4):111-114,4.基金项目
广州大学松田学院2020 年度科研规划立项项目(Gzdxstxy2020-12) （Gzdxstxy2020-12）
广州应用科技学院 2022年度校级教学质量与教学革工程项目(2022ZG008) （2022ZG008）
