基于深度学习的疲劳驾驶状态检测方法

A Method of Fatigue Driving State Detection Based on Deep Learning

目前疲劳驾驶检测算法大多基于单一的人工提取疲劳状态特征实现,且大部分算法结构复杂、鲁棒性低。为此,文章提出一种基于深度学习的疲劳检测方法,它采用卷积神经网络和Landmark算法来实现人脸图像特征点的自动提取,并使用SVM算法对疲劳特征进行分类,最后基于Perclos算法实现视频流图像的疲劳状态检测。实验结果表明,该方法能较好地提取疲劳特征,实现实时疲劳检测,且检测精度达96.8%。

Current domestic and overseas fatigue-recognition algorithms are implemented using fatigue features which are mostly singular and man-made.Most of those algorithms have complex structure,low efficiency and weak adaptability in face of driver's individual behavior.habit.To this end,this paper put forward a fatigue recognition algorithm based on deep learning.Firstly,the face image feature points are automatic extracted using convolutional neural network and landmark algorithm.Then the SVM algorithm is used to classify the fatigue characteristics.Finally,the fatigue state of the video stream image is detected based on the Perclos algorithm: The experimental results show that this method can obtain good fatigue characteristics,realize real-time fatigue detection,and its detection accuracy is 96.8%.