摘要
交通标志是确保道路安全的重要措施,对其检测是智能交通系统中的关键环节。文章引入YOLOv8网络模型对道路两旁的交通标志进行检测,并提出了一种基于Clo-YOLO的交通标志检测算法。在融合深层特征和浅层特征的过程中,引入了CloFormer模块,以增强目标的特征表达能力,并提高在目标被遮挡时的检测能力。与原模型的检测结果相比,Clo-YOLO模型在检测道路旁交通标志的任务中表现更为出色。
Traffic signs are important measures to ensure road safety,and their detection is a key link in intelligent transportation systems.The article introduces the YOLOv8 network model to detect traffic signs on both sides of the road and proposes a traffic sign detection algorithm based on Clo YOLO.In the process of integrating deep and shallow features,the CloFormer module is introduced to enhance the feature expression ability of the target and improve the detection ability when the target is occluded.Compared with the detection results of the original model,the Clo-YOLO model performs better in the task of detecting roadside traffic signs.
作者
杨宵
吕国
YANG Xiao;LV Guo(Hebei University of Architecture,Zhangjiakou,Hebei 075000,China)