机电工程技术2023，Vol.52Issue(12)：27-30,4.DOI:10.3969/j.issn.1009-9492.2023.12.007
基于深度学习的驾驶员分心驾驶行为预警算法
Driver Distracted Driving Behavior Warning Algorithm Based on Deep Learning
摘要
Abstract
Driving a car safely is a complex task that requires the driver's full attention.In order to improve drivers'attention,a set of deep learning-based early warning algorithm for driver distraction driving behavior is constructed,which can be used to achieve accurate monitoring of driver attention in all directions.In terms of facial fatigue detection for drivers,Dlib is used for facial key point detection.Then,the degree of eye and mouth opening and closing is calculated to determine whether to close or yawn,and the Perclos model is used to calculate the level of driver fatigue.Secondly,the detection of driver distraction behavior uses convolutional neural network YOLOv5 to detect whether the driver has distracted actions such as playing with their phone,smoking,and drinking water.The experimental results show that the algorithm has extremely high accuracy in detecting facial fatigue and driver distraction behavior when drivers are distracted by diversity,and it can give warnings,which helps to improve the safety of car driving.关键词
分心驾驶/卷积神经网络/预警算法Key words
distracted driving/convolutional neural network/early warning algorithm分类
信息技术与安全科学引用本文复制引用
欧阳壮,朱天军,文浩..基于深度学习的驾驶员分心驾驶行为预警算法[J].机电工程技术,2023,52(12):27-30,4.基金项目
广东省教育厅特色创新项目资助(2022KTSCX146) （2022KTSCX146）
广东省教育厅重点领域项目资助(2021ZDZX1061) （2021ZDZX1061）
