TY - GEN
T1 - Driver behavior recognition based on deep convolutional neural networks
AU - Yan, Shiyang
AU - Teng, Yuxuan
AU - Smith, Jeremy S.
AU - Zhang, Bailing
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2016/10/19
Y1 - 2016/10/19
N2 - Traffic safety is a severe problem around the world. Many road accidents are normally related with the driver's unsafe driving behavior, e.g. eating while driving. In this work, we propose a vision-based solution to recognize the driver's behavior based on convolutional neural networks. Specifically, given an image, skin-like regions are extracted by Gaussian Mixture Model, which are passed to a deep convolutional neural networks model, namely RCNN, to generate action labels. The skin-like regions are able to provide abundant semantic information with sufficient discriminative capability. Also, RCNN is able to select the most informative regions from candidates to facilitate the final action recognition. We tested the proposed methods on Southeast University Driving-posture Dataset and achieve mean Average Precision(mAP) of 97.76% on the dataset which prove the proposed method is effective in drivers's action recognition.
AB - Traffic safety is a severe problem around the world. Many road accidents are normally related with the driver's unsafe driving behavior, e.g. eating while driving. In this work, we propose a vision-based solution to recognize the driver's behavior based on convolutional neural networks. Specifically, given an image, skin-like regions are extracted by Gaussian Mixture Model, which are passed to a deep convolutional neural networks model, namely RCNN, to generate action labels. The skin-like regions are able to provide abundant semantic information with sufficient discriminative capability. Also, RCNN is able to select the most informative regions from candidates to facilitate the final action recognition. We tested the proposed methods on Southeast University Driving-posture Dataset and achieve mean Average Precision(mAP) of 97.76% on the dataset which prove the proposed method is effective in drivers's action recognition.
KW - Convolutional Neural Networks
KW - Driver's Behavior Recognition
KW - Gaussian Mixture Model
KW - RCNN
KW - Skin-color Modeling
UR - https://www.scopus.com/pages/publications/84997706096
U2 - 10.1109/FSKD.2016.7603248
DO - 10.1109/FSKD.2016.7603248
M3 - Conference Proceeding
AN - SCOPUS:84997706096
T3 - 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
SP - 636
EP - 641
BT - 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
A2 - Du, Jiayi
A2 - Liu, Chubo
A2 - Li, Kenli
A2 - Wang, Lipo
A2 - Tong, Zhao
A2 - Li, Maozhen
A2 - Xiong, Ning
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2016
Y2 - 13 August 2016 through 15 August 2016
ER -