The motor imagery-based brain-computer interface (BCI) using electroencephalography (EEG) has been receiving attention from neural engineering researchers and is being applied to various rehabilitation applications. However, the performance degradation caused by motor imagery EEG with very low single-to-noise ratio faces several application issues with the use of a BCI system. In this paper, we propose a novel motor imagery classification scheme based on the continuous wavelet transform and the convolutional neural network. Continuous wavelet transform with three mother wavelets is used to capture a highly informative EEG image by combining time-frequency and electrode location. A convolutional neural network is then designed to both classify motor imagery tasks and reduce computation complexity. The proposed method was validated using two public BCI datasets, BCI competition IV dataset 2b and BCI competition II dataset III. The proposed methods were found to achieve improved classification performance compared with the existing methods, thus showcasing the feasibility of motor imagery BCI.

Among various methods for frequency recognition of the steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) study, a task-related component analysis (TRCA), which extracts discriminative spatial filters for classifying electroencephalogram (EEG) signals, has gathered much interest. The TRCA-based SSVEP method yields lower computational cost and higher classification performance compared to existing SSVEP methods. In spite of its utility, the TRCA-based SSVEP method still suffers from the degradation of the frequency recognition rate in cases where EEG signals with a short length window are used. To address this issue, here, we propose an improved strategy for decoding SSVEPs, which is insensitive to a window length by carrying out two-step TRCA. The proposed method reuses the spatial filters corresponding to target frequencies generated by the TRCA. Followingly, the proposed method accentuates features for target frequencies by correlating individual template and test data. For the evaluation of the performance of the proposed method, we used a benchmark dataset with 35 subjects and confirmed significantly improved performance comparing with other existing SSVEP methods. These results imply the suitability as an efficient frequency recognition strategy for SSVEP-based BCI applications.