基于卷积神经网络的驾驶人行为识别方法研究

X. U. Dan; D. A.I. Yong; J. I. Junhong

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

Translated title of the contribution: Research on driver behavior recognition method based on convolutional neural network

X. U. Dan
, D. A.I. Yong
, J. I. Junhong

School of Mechatronics Engineering, Harbin Institute of Technology

Research output: Contribution to journal › Article › peer-review

11 Scopus citations

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.

Translated title of the contribution	Research on driver behavior recognition method based on convolutional neural network
Original language	Chinese (Traditional)
Pages (from-to)	12-17
Number of pages	6
Journal	China Safety Science Journal
Volume	29
Issue number	10
DOIs	https://doi.org/10.16265/j.cnki.issn1003-3033.2019.10.003
State	Published - Oct 2019
Externally published	Yes

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

10.16265/j.cnki.issn1003-3033.2019.10.003

Cite this

@article{e01c104af1b3456f846782e5dbcaecb7,

title = "基于卷积神经网络的驾驶人行为识别方法研究",

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.",

keywords = "back propagation, convolutional neural network (CNN), deep learning, driver behavior recognition, driving safety, forward propagation",

author = "Dan, \{X. U.\} and Yong, \{D. A.I.\} and Junhong, \{J. I.\}",

year = "2019",

month = oct,

doi = "10.16265/j.cnki.issn1003-3033.2019.10.003",

language = "繁体中文",

volume = "29",

pages = "12--17",

journal = "China Safety Science Journal",

issn = "1003-3033",

publisher = "Editorial Department of China Safety Science Journal ",

number = "10",

}

TY - JOUR

T1 - 基于卷积神经网络的驾驶人行为识别方法研究

AU - Dan, X. U.

AU - Yong, D. A.I.

AU - Junhong, J. I.

PY - 2019/10

Y1 - 2019/10

N2 - 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.

AB - 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.

KW - back propagation

KW - convolutional neural network (CNN)

KW - deep learning

KW - driver behavior recognition

KW - driving safety

KW - forward propagation

UR - https://www.scopus.com/pages/publications/85109372014

U2 - 10.16265/j.cnki.issn1003-3033.2019.10.003

DO - 10.16265/j.cnki.issn1003-3033.2019.10.003

M3 - 文章

AN - SCOPUS:85109372014

SN - 1003-3033

VL - 29

SP - 12

EP - 17

JO - China Safety Science Journal

JF - China Safety Science Journal

IS - 10

ER -

基于卷积神经网络的驾驶人行为识别方法研究

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this