基于YOLOv8的车辆驾驶员疲劳检测应用研究  

Application of Vehicle Driver Fatigue DetectionBased on YOLOv8

作　　者：王睿 WANG Rui(College of Industrial Education,Technology University of the Philippines,Manila 0900,Philippines;Department of Information and Artificial Intelligence,Anhui Business College,Wuhu 241002,China)

机构地区：[1]菲律宾科技大学工业教育学院,菲律宾马尼拉0900 [2]安徽商贸职业技术学院信息与人工智能学院,安徽芜湖241002

出　　处：《安徽工程大学学报》2024年第5期26-31,共6页Journal of Anhui Polytechnic University

基　　金：安徽省高校优秀人才支持计划重点项目(gxyqZD2020056);安徽省高校自然科学重点项目(2022AH052741);安徽商贸职业技术学院技术技能创新服务平台项目(2022ZDG01);安徽商贸职业技术学院“双高计划”项目(2020sgxm05-4)。

摘　　要：疲劳驾驶是导致交通事故发生的主要原因,由于疲劳驾驶检测场景的复杂性和实时性要求,提出了一种基于YOLOv8的车辆驾驶员疲劳检测预警设计方法,对YOLOv8算法在注意力机制、数据增强、轻量化网络等方面进行改进,提高车辆驾驶员疲劳检测的识别精度和检测速率,同时提取人脸关键点计算眼睛纵横比(EAR),建立疲劳度评价分类模型,实现对疲劳驾驶的综合判断和预警。搭建车辆驾驶员疲劳检测实验平台,并进行验证。结果表明,该方法可以准确获取疲劳检测结果,准确率达到94%。Fatigue driving is the main cause of traffic accidents.Due to the complexity and real-time requirements of fatigue driving detection scenarios,a vehicle driver fatigue detection and warning design method based on YOLOv8 is proposed.The YOLOv8 algorithm is improved in attention mechanism,data augmentation,lightweight network,etc.to improve the recognition accuracy and detection rate of vehicle driver fatigue detection.Mean while,key facial points are extracted to calculate the eye aspect ratio(EAR),and a fatigue evaluation classification model is established to achieve comprehensive judgment and warning of fatigue driving.A vehicle driver fatigue detection experimental platform is built to verify it.The results show that this approach can accurately obtain fatigue detection results,with an accuracy rate of 94%.

关 键 词：YOLOv8 疲劳检测 注意力机制 眼睛纵横比 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]