Driver Activity Recognition for Intelligent Vehicles: A Deep Learning Approach

doi:10.1109/TVT.2019.2908425

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 平行智能技术与系统团队

	Driver Activity Recognition for Intelligent Vehicles: A Deep Learning Approach
	Xing, Yang 1; Lv, Chen 1; Wang, Huaji 2; Cao, Dongpu 3; Velenis, Efstathios 4; Wang, Fei-Yue5
发表期刊	IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
ISSN	0018-9545
	2019-06-01
卷号	68 期号:6 页码:5379-5390
通讯作者	Lv, Chen(lyuchen@ntu.edu.sg) ; Cao, Dongpu(dongpu.cao@uwaterloo.ca)
摘要	Driver decisions and behaviors are essential factors that can affect the driving safety. To understand the driver behaviors, a driver activities recognition system is designed based on the deep convolutional neural networks (CNN) in this paper. Specifically, seven common driving activities are identified, which are the normal driving, right mirror checking, rear mirror checking, left mirror checking, using in-vehicle radio device, texting, and answering the mobile phone, respectively. Among these activities, the first four are regarded as normal driving tasks, while the rest three are classified into the distraction group. The experimental images are collected using a low-cost camera, and ten drivers are involved in the naturalistic data collection. The raw images are segmented using the Gaussian mixture model to extract the driver body from the background before training the behavior recognition CNN model. To reduce the training cost, transfer learning method is applied to fine tune the pre-trained CNN models. Three different pre-trained CNN models, namely, AlexNet, GoogLeNet, and ResNet50 are adopted and evaluated. The detection results for the seven tasks achieved an average of 81.6% accuracy using the AlexNet, 78.6% and 74.9% accuracy using the GoogLeNet and ResNet50, respectively. Then, the CNN models are trained for the binary classification task and identify whether the driver is being distracted or not. The binary detection rate achieved 91.4% accuracy, which shows the advantages of using the proposed deep learning approach. Finally, the real-world application are analyzed and discussed.
关键词	Driver behavior driver distraction convolutional neural network transfer learning
DOI	10.1109/TVT.2019.2908425
关键词[WOS]	BEHAVIOR
收录类别	SCI
语种	英语
资助项目	Young Elite Scientist Sponsorship Program by CAST[2017QNRC001] ; SUG-NAP of Nanyang Technological University, Singapore[M4082268.050] ; Young Elite Scientist Sponsorship Program by CAST[2017QNRC001] ; SUG-NAP of Nanyang Technological University, Singapore[M4082268.050]
项目资助者	Young Elite Scientist Sponsorship Program by CAST ; SUG-NAP of Nanyang Technological University, Singapore
WOS研究方向	Engineering ; Telecommunications ; Transportation
WOS类目	Engineering, Electrical & Electronic ; Telecommunications ; Transportation Science & Technology
WOS记录号	WOS:000472563200016
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：196[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/26049
专题	多模态人工智能系统全国重点实验室_平行智能技术与系统团队
通讯作者	Lv, Chen; Cao, Dongpu
作者单位	1.Nanyang Technol Univ, Sch Mech & Aerosp Engn, Singapore 639798, Singapore 2.AVL Powertrain UK Ltd, Coventry CV4 7EZ, W Midlands, England 3.Univ Waterloo, Dept Mech & Mechatronics Engn, Waterloo, ON N2L 3G1, Canada 4.Cranfield Univ, Adv Vehicle Engn Ctr, Bedford MK43 0AL, England 5.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Xing, Yang,Lv, Chen,Wang, Huaji,et al. Driver Activity Recognition for Intelligent Vehicles: A Deep Learning Approach[J]. IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY,2019,68(6):5379-5390.
APA	Xing, Yang,Lv, Chen,Wang, Huaji,Cao, Dongpu,Velenis, Efstathios,&Wang, Fei-Yue.(2019).Driver Activity Recognition for Intelligent Vehicles: A Deep Learning Approach.IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY,68(6),5379-5390.
MLA	Xing, Yang,et al."Driver Activity Recognition for Intelligent Vehicles: A Deep Learning Approach".IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 68.6(2019):5379-5390.