Multi-recognition algorithms of human's mental fatigue state based on EEG

Abstract

In this paper, the EEG was respectively analyzed by the algorithms of power spectrum and wavelet entropy. And both of the algorithms are based on the same experiment. In addition, through the BP neural network, the state of mental fatigue for human being can be analyzed quickly and correctly. By experiment, when the value of δ is higher than before, at the same time, the value of β, α and θ are lower, or the average value of wavelet entropy is lower, the state of mental fatigue can be sure. Finally, the result, wavelet entropy is better in saving time and accuracy, was got from comparing the two algorithms.