Driver drowsiness recognition via transferred deep 3D convolutional network and state probability vector
- PDF / 867,922 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 59 Downloads / 208 Views
Driver drowsiness recognition via transferred deep 3D convolutional network and state probability vector Lei Zhao 1
2
3
& Zengcai Wang & Guoxin Zhang & Huanbing Gao
4
Received: 3 August 2019 / Revised: 2 May 2020 / Accepted: 24 June 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Driver drowsiness is a major cause of road accidents. In this study, a novel approach that detects human drowsiness is proposed and investigated. First, driver face and facial landmarks are detected to extract facial region from each frame in a video. Then, a residual-based deep 3D convolution neural network (CNN) that learned from an irrelevant dataset is constructed to classify driver facial image sequences with a certain number of frames for obtaining its drowsiness output probability value. After that, a certain number of output probability values is concatenated to obtain the state probability vector of a video. Finally, a recurrent neural network is adopted to classify constructed probability vector and obtain the recognition result of driver drowsiness. The proposed method is tested and investigated using a public drowsy driver dataset. Experimental results demonstrate that similar to 2D CNN, 3D CNN can learn spatiotemporal features from irrelevant dataset to improve its performance obviously in driver drowsiness classification. Furthermore, the proposed method performs stably and robustly, and it can achieve an average accuracy of 88.6%. Keywords Driver drowsiness detection . 3D convolution neural network . State probability vector . Transfer learning . Residual learning
1 Introduction Driver fatigue is a major cause of accidents. Compared with drunk driving, overspeeding, and other risky driving behaviors, driver drowsiness is difficult to detect and prevent, thereby * Lei Zhao [email protected] Zengcai Wang [email protected] Guoxin Zhang [email protected] Huanbing Gao [email protected] Extended author information available on the last page of the article
Multimedia Tools and Applications
compromising the security and safety of drivers, passengers, and pedestrians. In recent years, with the development of artificial intelligence technology, autonomous driving technology has progressed rapidly. However, road vehicles still need drivers. Furthermore, with conditional autonomous driving systems, driver should still be ready to take over if these systems do not operate correctly [1]. Therefore, research on effective methods for detecting driver drowsiness is important to improve transport safety. Detection methods can be divided into three categories according to basis: driver physiological parameters, operational behavior, and physical conditions [2]. Physiological signals change evidently when drivers feel drowsy. In recent years, many physiological signal detection methods for recognizing driver drowsiness have been studied [3]. Physiological signal tests include electroencephalogram, electrocardiogram, electrooculogram, and electromyogram [4]. Physiology-based meth
Data Loading...
