Open AccessDriver Distraction Behavior Detection Method Based on Deep Learning
Author(s) -
Peng Mao,
Kunlun Zhang,
Liang Da
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/782/2/022012
Subject(s) - distraction , distracted driving , computer science , deep learning , phone , process (computing) , face (sociological concept) , face detection , artificial intelligence , real time computing , simulation , feature extraction , psychology , facial recognition system , cognitive psychology , social science , linguistics , philosophy , sociology , operating system
With the rapid development of road traffic in China, driver safety accidents caused by road traffic accidents are increasing year by year. According to statistics of relevant departments, 20%-30% of traffic safety accidents are caused by distracted behaviors of drivers. For this reason, this paper proposes a driver distraction behavior detection method based on deep learning, which uses PCN and DSST algorithms for face detection, location and dynamic face tracking. Finally, YOLOV3 object detection algorithm is used to identify distracting behaviors such as smoking and making phone calls around a person’s face. The method can detect distracted behaviors in the driving process in real time and has high detection accuracy.