A Method of Motor Imagery EEG Recognition Based on CNN-ELM

Abstract

It is the key of brain-computer interface technology to extract electroencephalogram (EEG) data features effectively and classify them accurately. In view of the characteristics of non-stationarity and obvious time-frequency characteristics of motor imagery EEG signals, this paper proposes a method for recognition of motor imagery EEG signals based on S-transform time-frequency image combined with convolutional neural network (CNN) and extreme learning machine (ELM). In the BCI competition dataset, firstly, the S-transform time-frequency image of C3 and C4 electrode signals is obtained, and then the characteristic frequency bands are extracted from the time-frequency image for combination. Finally, the combined image is used as the input of neural network to realize the recognition of left-right hand motor imagery EEG signals. Experimental results show that this method is superior to the ordinary convolutional neural network.