EEG-based Motor Imagery Classification through Transfer Learning of the CNN

Samaneh Taheri, Mehdi Ezoji
The 11th Iranian and the first International Conference on Machine Vision and Image Processing (MVIP 2020)

Abstract

Brain computer interface (BCI) is a system which is able to translate EEG signals into comprehensive commands for the computers. EEG-basedmotor imagery (MI) signals are One of the most widely used signals in this topic.In this paper, an efficient algorithm to classify 2-class MI signals based on theconvolutional neural network (CNN) through the transfer learning. To thisend, different 3D representations of EEG signals are injected into the CNN.These proposed 3D representations are prepared by combination of some frequency and time-frequency algorithms such as Fourier Transform, CSP, DCTand EMD. Then, CNN will be trained to classify MI-EEG signals. The averageaccuracy of classification for 5 subjects achieved 98.5% on the BCI competitioniii database IVa.

Keywords: Motor Imagery, EEG Signal, Convolutional Neural Network, Transfer Learning