基于YOLOv8的自动驾驶车道检测改进算法  

Improved Curve Lane Detection Algorithm for Autonomous Vehicles Based on YOLOv8

作　　者：阮庭海 樊卫华[1] RUAN Tinghai;FAN Weihua(School of Automation,Nanjing University of Science&Technology,Nanjing 210094,China)

机构地区：[1]南京理工大学自动化学院,南京210094

出　　处：《计算机测量与控制》2025年第2期31-36,共6页Computer Measurement &Control

基　　金：江苏省科技重大专项(BG2024041)。

摘　　要：针对自动驾驶领域的车道自动检测中存在的检测准确率低、实际应用难等问题,研究基于YOLO与传统图像处理算法混合的车道检测算法;基于车载传感器拍摄的视频,利用YOLOv8算法检测并标记车前/侧方附近的物体,并将图像视角转换到鸟瞰视角,利用基于滑动窗口的二次多项式法识别当前帧的车道线,融合前序帧的车道信息,检测出当前帧的车道;经过数据集和实际场景的测试表明,算法的检测准确性提升10%以上,检测速度明显提升。Aiming at the problems of low detection accuracy and difficult practical application in automatic lane detection in the field of automatic driving,a mixed lane detection algorithm based on YOLO and traditional image processing algorithms is studied.Based on the video captured by the on-board sensors,YOLOv8 algorithm is used to detect and mark the objects near the front or side of the vehicle,and the image viewpoint is converted to a bird s-eye view,and the quadratic polynomial method based on sliding window is used to identify the lane lines with current frame,and fuse the lane information of previous frames to detect the lane in the current frame.Tests on datasets and actual scenarios show that the algorithm improves the detection accuracy by more than 10%and significantly increases the detection speed.

关 键 词：YOLO 曲线车道检测 滑动窗口 自动驾驶 计算机视觉 

分 类 号：TP242.3[自动化与计算机技术—检测技术与自动化装置]