Open AccessFatigue detection based on multi-feature fusion of fatigue behavior
Author(s) -
Xing Chen,
Lumei Su,
Qin Meixin
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/853/1/012056
Subject(s) - computer science , feature (linguistics) , matlab , mental fatigue , artificial intelligence , fatigue testing , pattern recognition (psychology) , engineering , structural engineering , psychology , philosophy , linguistics , applied psychology , operating system
The research of fatigue driving detection is of great significance due to the contribution in fatigue driving prevention. However, most of current fatigue detection study are limited to unavailable real-time performance, unsatisfied accuracy rate, rough fatigue level and complicated implementation. In this paper, a novel method based on multi fatigue behavior features are proposed to improve the accuracy of fatigue detection. First of all, multi facial behavior features including eyes, mouth and head features are extracted. And then a new multi-feature fusion fatigue detection approach based on specific fatigue behavior feature anaylsis is proposed to recognize the previously extracted features. For better accuracy and applicability, a more detailed fatigue levels are newly used to subdivide fatigue behavior into three degree of fatigue. The experimental results based Matlab simulation and Android App indicate that the proposed method could accurately and real-time detect whether the tester is in a state of fatigue driving, and also determine the fatigue level.