Driver drowsiness detection using facial dynamic fusion information and a DBN. Issue 2 (6th December 2017)

• Record Type: Journal Article
• Authors: Zhao, Lei; Wang, Zengcai; Wang, Xiaojin; Liu, Qing
• Is Part Of: IET intelligent transport systems. Volume 12:Issue 2(2018)
• Page Start: 127
• Page End: 133
• Publication Date: 2017-12-06
• Subjects: driver information systems -- road traffic -- road safety -- feature extraction -- emotion recognition -- face recognition -- object detection -- cameras -- video signal processing -- belief networks -- image fusion; driver drowsiness detection methods -- facial dynamic fusion information -- DBN -- deep belief network -- traffic accidents -- road traffic safety improvement -- driver drowsiness expression recognition -- landmark extraction -- texture extraction -- high‐definition camera -- facial drowsiness expression classification -- driver drowsiness dataset -- genders -- ages -- head poses -- illuminations -- facial subregions -- temporal resolutions -- driver fatigue recognition; Intelligent transportation systems -- Periodicals; Electronics in transportation -- Periodicals; 388.31205
• DOI: 10.1049/iet-its.2017.0183
• Languages: English
• ISSNs: 1751-956X

Abstract

Abstract : Driver drowsiness is a frequent cause of traffic accidents. Research on driver drowsiness detection methods is important to improve road traffic safety. Previous driving fatigue detection methods frequently extracted single features such as eye or mouth changes and trained shallow classifiers, which limit the generalisation capability of these methods. This study proposes a framework for recognising driver drowsiness expression by using facial dynamic fusion information and a deep belief network (DBN) to address the aforementioned problem. First, the landmarks and textures of the facial region are extracted from videos captured using a high‐definition camera. Then, a DBN is built to classify facial drowsiness expressions. Finally, the authors' method is tested on a driver drowsiness dataset, which includes different genders, ages, head poses and illuminations. Certain experiments are also carried out to investigate the effects of different facial subregions and temporal resolutions on the accuracy of driver fatigue recognition. Results demonstrate the validity of the proposed method, which has an average accuracy of 96.7%.