Classifying Motor Imagery EEG Signals Using the Deep Residual Network

Abstract

Classification of motor imagery electroencephalogram (EEG) signals is widely applied in the non-invasive brain–computer interface (BCI) field. Studying the feature and correlations is important for the classification of motor imagery. In this paper, we propose a method based on Deep Residual Network (DRN) for classifying the EEG signals of motor imagery. Firstly, the raw EEG data are represented in the time-frequency field using continuous wavelet transformation, and then a DRN model is built to classify MI tasks. Results show improved accuracy of the proposed method compared with other classification methods (SCP + SVM).