As driving population and the distribution rate of cars increase continuously, the highway plays an important role in the boost of the national economic growth, the vitalization of local economy, the enhancement of the convenience in people’s lives and the unity of people through interactions between regions. Even though the rate of car accidents in the highway has decreased, according to the statistics of Road Traffic Authority in 2010, the rate of car accidents caused by dozing while driving stays as the same as on average and the death toll consists of 40% of all, which costs 32.7 billion won (in 2010) a year.
For this reason, drowsiness detection and warning system for drivers has recently become a very important issue. Monitoring physiological signals provides the possibility of detecting features of drowsiness and fatigue of drivers. Many papers have been published that to measure electroencephalogram(EEG) signals is the effective way in order to be aware of fatigue and drowsiness of drivers. The aim of this study is to extract drowsiness-related features from a set of EEG signals and to classify the features into three states: alertness, transition, and drowsiness. This paper proposes a drowsiness detection system using errors-in-variables(EIV) for extraction of feature vectors and multilayer perceptron (MLP) for classification. The proposed method evaluates robustness for noise and compares to the previous one using linear predictive coding (LPC) combined with MLP. From evaluation results, we conclude that the proposed scheme outperforms the previous one in the low signal-to-noise ratio regime.
Since electroencephalogram(EEG) has non-linear and non-stationary properties, it is effective to analyze the characteristic of EEG with time-frequency method rather than spectrum method. In this letter, we propose the modified drowsiness detection system using discrete wavelet transform combined with errors-in-variables and multilayer perceptron methods. For the comparison of the proposed scheme with the previous one, the state ‘ambiguousness’ is added to the previous states of drivers: ‘alertness,’ ‘transition,’ and ‘drowsiness.’ From the computer simulation using machine learning, we confirm that the proposed scheme outperforms the previous one for some conditions.
To prevent the accident from happening, the constant effort for the expansion of a shelter for sleeping and a campaign has a selective and temporary effect on drivers, but this does not fundamentally help to solve the problem. Safety facilities to stimulate drivers’ hearing and touching senses were installed with the limit of its distance in-between and of its effectiveness, so their practicality is declining.
The second purpose of this paper is development of human arousal inducing interface using steady-state visual evoked potential (SSVEP) and its verification through experiments. In order to develop the model, computer-based SSVEP program simulation is preliminary performed. From the results of the simulation, stimulus pattern is decided to checkerboard and SSVEP frequency range is set into beta wave(13~30Hz). 11 subjects of 14 subjects has meaningful results so that the proposed system has an effect on inducing driver’s alertness. Therefore, this research can be a previous study suggesting experimental verification and standard model considering the development of road facilities inducing driver’s alertness