Developing and evaluating a mobile driver fatigue detection network based on electroencephalograph signals

Abstract

The rapid development of driver fatigue detection technology indicates important significance of traffic safety. The authors’ main goals of this Letter are principally three: (i) A middleware architecture, defined as process unit (PU), which can communicate with personal electroencephalography (EEG) node (PEN) and cloud server (CS). The PU receives EEG signals from PEN, recognises the fatigue state of the driver, and transfer this information to CS. The CS sends notification messages to the surrounding vehicles. (ii) An android application for fatigue detection is built. The application can be used for the driver to detect the state of his/her fatigue based on EEG signals, and warn neighbourhood vehicles. (iii) The detection algorithm for driver fatigue is applied based on fuzzy entropy. The idea of 10-fold cross-validation and support vector machine are used for classified calculation. Experimental results show that the average accurate rate of detecting driver fatigue is about 95%, which implying that the algorithm is validity in detecting state of driver fatigue.

Keywords

• electroencephalography
• road traffic
• cloud computing
• fuzzy logic
• entropy
• support vector machines
• accident prevention
• medical signal detection
• mobile driver fatigue detection network
• electroencephalograph signals
• traffic safety
• middleware architecture
• process unit
• personal electroencephalography node
• cloud server
• android application
• fuzzy entropy
• support vector machine