Estimate vigilance level in driving simulation based on sparse representation

Avoiding fatal accidents caused by low vigilance level in driving is very important in our daily lives. Electroencephalography(EEG) has been proved very effective for measuring the level of vigilance. In this paper, we distinguish vigilance level into three classes which are ´alert´, ´fatigue´ and ´sleeping´ by using sparse representation classification(SRC). Six features from each frequency band are got from samples of EEG data. Random feature is used to reduce the dimension of features. Actually there is almost no training process before the classification. The accuracy in classification of three classes reaches about 90% on average.