基于卷积神经网络的驾驶人行为识别方法研究

doi:10.16265/j.cnki.issn1003-3033.2019.10.003

中国安全科学学报 ›› 2019, Vol. 29 ›› Issue (10): 12-17.doi: 10.16265/j.cnki.issn1003-3033.2019.10.003

基于卷积神经网络的驾驶人行为识别方法研究

徐丹, 代勇, 纪军红副教授

哈尔滨工业大学机电工程学院,黑龙江哈尔滨 150001

收稿日期:2019-07-13 修回日期:2019-09-04 出版日期:2019-10-28 发布日期:2020-10-27
作者简介:徐丹 (1996—),女,黑龙江大庆人,硕士研究生,主要研究方向为数据分析与智能图像处理。E-mail: xudan_hit@163.com。
基金资助:
机器人技术与系统国家重点实验室自主课题(SKLRS201705A)。

Research on driver behavior recognition method based on convolutional neural network

XU Dan, DAI Yong, JI Junhong

School of Mechatronics Engineering, Harbin Institute of Technology, Harbin Heilongjiang 150001, China

Received:2019-07-13 Revised:2019-09-04 Online:2019-10-28 Published:2020-10-27

摘要/Abstract

摘要： 为探究汽车驾驶人非安全驾驶行为的识别问题,在简要分析现有驾驶人行为识别方法的基础上,提出一种基于卷积神经网络(CNN)的驾驶人行为识别方法,分析CNN前向传播与反向传播过程,给出处理驾驶人行为识别问题的CNN网络架构。结果表明:用该方法可识别,其平均识别率达97.13%,相对于传统提取方向梯度直方图特征(HOG),并用随机森林(RF)分类的算法,该方法的识别率平均提高了3.62%。

关键词: 驾驶人行为识别, 卷积神经网络(CNN), 前向传播, 反向传播, 深度学习, 驾驶安全

Abstract: In order to explore identification of unsafe driving behaviors of car drivers, concrete studies were carried out on CNN-based driver behavior recognition algorithm building on brief analysis of existing driver behavior recognition methods. CNN forward propagation and back propagation processes were explored and a CNN network architecture that deals with driver behavior recognition was presented. The results show that this method achieves an average recognition rate of 97.13% on state-farm driver behavior dataset, and compared with traditional algorithm, it has improved 3.62% on average in extracting histogram of oriented gradient(HOG) feature and using random forest(RF) classification for identification.

Key words: driver behavior recognition, convolutional neural network (CNN), forward propagation, back propagation, deep learning, driving safety

中图分类号:

X910

徐丹, 代勇, 纪军红. 基于卷积神经网络的驾驶人行为识别方法研究[J]. 中国安全科学学报, 2019, 29(10): 12-17.

XU Dan, DAI Yong, JI Junhong. Research on driver behavior recognition method based on convolutional neural network[J]. China Safety Science Journal, 2019, 29(10): 12-17.

参考文献

[1] 中华人民共和国公安部交通管理局. 2018年全国小汽车保有量首次突破2亿辆 [EB/OL]. (2019-01-12). http://www.mps.gov.cn/n2254098/n4904352/c6354939/content.html.
[2] WANG Zhongmin, ZHANG Yao, HENG Xia. Driving behavior recognition based on sparse filtering-convolutional neural network[J]. Computer Engineering and Applications, 2018, 54(11): 128-132.
[3] 杜勇,王春明,崔金,等. 基于稀疏时空特征描述的驾驶者多种非安全驾驶行为识别[J]. 智能计算机与应用, 2018, 8(6): 49-53.
DU Yong, WANG Cunming, CUI Jin, et al. Recognition of driver specific unsafe driving behaviors based on sparse spatio-temporal feature description[J]. Intelligent Computer and Applications, 2018, 8(6): 49-53.
[4] GURUPACKIAM S, JONES S, TURNER D. Characterization of arterial traffic congestion through analysis of operational parameters (Gap acceptance and lane changing) [J]. Lane Changing, 2010, 5(1): 1-35.
[5] LIU Ronghui, VANVLIET D, WATLING D. Microsimulation models incorporating both demand and supply dynamics[J]. Transportation Research Part A: Policy and Practice, 2006, 40 (2): 125-150.
[6] 楚文慧,吴超仲,张晖,等. 基于个性化行为模型的驾驶疲劳识别方法[J]. 中国安全科学学报, 2018,28(6): 43-48.
CHU Wenhui, WU Chaozhong, ZHANG Hui, et al. Driver behavior model and its application in driver fatigue identification [J]. China Safety Science Journal, 2018,28(6): 43-48.
[7] 张日东. 出租车司机吸烟行为评判标准及自动检测算法研究[D]. 北京:北京工业大学, 2018.
ZHANG Ridong. Study on the evaluation standard of smoking behavior of taxi drivers and automatic detection algorithm[D]. Beijing:Beijing University of Technology, 2018.
[8] 周雯,吕晓军,程清波, 等.基于自适应特征聚类的机车司机驾驶行为检测算法研究[C]. 第九届中国智能交通年会论文集, 2014:1 191.
ZHOU Wen, LYU Xiaojun, CHENG Qingbo, et al. Driving behavior floco motive detection algorithm based on adaptive feature clustering[C]. Proceedings of the 9^th China Intelligent Transportation Annual Conference, 2014: 1191.
[9] 邓院昌,史晨军.基于贝叶斯结构方程模型的疲劳驾驶行为意图研究[J].安全与环境学报,2019,19(2): 520-526.
DENG Yuanchang, SHI Chenjun. Psycho-intentional analysis for the factors leading to fatigue driving based on the Bayesian-SEM [J]. Journal of Safety and Environment, 2019, 19(2): 520-526
[10] KRIZHEVSKY A, SUTSKEVER I, HINTON G. ImageNet classification with deep convolutional neural networks [J]. Advances in Neural Information Processing Systems (NIPS), 2012, 25(2): 1 097-1 105.
[11] ROSS G, JEFF D, TREVOR D, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
[12] ABDEL O, MOHAMED A, JIANG Hui, et al. Convolutional neural networks for speech recognition[J]. IEEE/ACM Transactions Audio, Speech, Language Process, 2014,22(10): 1 533-1 545.
[13] KAREN S, ANDREW Z. Transactions very deep convolutional networks for large-scale image recognition[C]. IEEE Conference on Computer Vision and Pattern Recognition, 2014: 1-14.
[14] SZEGEDY C, LIU Wei, JIA Yangqing, et al. Going deeper with convolutions[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 1-9.
[15] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification [C]. IEEE International Conference on Computer Vision (ICCV), 2014: 1 026-1 034.

基于卷积神经网络的驾驶人行为识别方法研究

Research on driver behavior recognition method based on convolutional neural network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张萌, 韩豫, 刘泽锋. 深度学习下建筑工人高空安全防护装备检测方法[J]. 中国安全科学学报, 2022, 32(5): 140-146.
[2]	郭玮, 袁宏永, 薛明, 魏平岩. SAR影像洪水淹没范围深度学习提取方法[J]. 中国安全科学学报, 2022, 32(4): 177-184.
[3]	乔洁, 徐鑫, 刘传攀, 赵俊玮, 刘永涛. 驾驶人危险感知能力影响因素及干预方式综述^*[J]. 中国安全科学学报, 2022, 32(2): 34-41.
[4]	何永勃, 曹祝兵. 基于IPSO-BP模型的火灾气体传感器气压补偿算法^*[J]. 中国安全科学学报, 2022, 32(2): 107-114.
[5]	石娟, 常丁懿, 郑鹏. 基于BP神经网络的建筑工人不安全行为预警模型^*[J]. 中国安全科学学报, 2022, 32(1): 27-33.
[6]	段照斌, 杜海龙, 张鹏. 基于QAR2Vec模型的QAR数据特征提取[J]. 中国安全科学学报, 2021, 31(1): 145-152.
[7]	赵欣然, 张琪, 王卫东, 徐志强. 可燃性粉尘云的图像检测方法[J]. 中国安全科学学报, 2020, 30(4): 8-13.
[8]	宋高峰, 张延兵, 孙培培, 沈硕勋, 王志荣. 大型储罐声发射技术下的安全评价方法[J]. 中国安全科学学报, 2020, 30(3): 60-66.
[9]	郝永梅, 杜璋昊, 杨文斌, 邢志祥, 蒋军成, 岳云飞. 基于改进ELMD和多尺度熵的管道泄漏信号识别[J]. 中国安全科学学报, 2019, 29(8): 105-111.
[10]	熊若鑫, 宋元斌, 王宇轩, 段彦娟. 基于CNN的3D姿势估计在建筑工人行为分析中的应用[J]. 中国安全科学学报, 2019, 29(7): 64-69.
[11]	宋博. DEA-BP神经网络下地铁车站深基坑施工安全评价[J]. 中国安全科学学报, 2019, 29(5): 91-96.
[12]	胡立伟, 何越人, 佘天毅, 孟玲, 杨锦青. 高速公路应急救援中心选址优化模型[J]. 中国安全科学学报, 2019, 29(5): 145-150.
[13]	汪洋, 王俊刚. 基于深度学习算法的铁路列车运行安全检测^*[J]. 中国安全科学学报, 2018, 28(S2): 41-45.
[14]	冯子健. 基于BP神经网络的铁路货运安全风险评价研究[J]. 中国安全科学学报, 2018, 28(S1): 178-185.
[15]	柴超君, 凌永生, 岳琪, 贾文宝. 维数约简用于BPNN的核事故源项估算方法[J]. 中国安全科学学报, 2018, 28(9): 122-127.