Ince-EEGNet: A Novel Fatigue Degree Detection Method Based on EEG

In this paper, we proposed a convolutional neural network based on EEGNet, which can be used to classify the degree of fatigue, without manual feature extraction, and directly take the original EEG as input, and the results of five-fold cross-validation show that the performance of the network is better than that of shallowConvNet and EEGNet, with the highest accuracy of 98.02%. Using EEG samples of different frequency bands as model inputs, it was found that the sample accuracy of gamma rhythm and beta rhythm was higher, both exceeding 90 %, and the accuracy of delta rhythm and theta rhythm was lower, only about 50 %. In addition, the frontal lobe had the highest classification accuracy in all brain regions, reaching nearly 94%, and the left brain had an accuracy of about 4 % higher than that of the right brain.