Multiple graph fusion based on Riemannian geometry for motor imagery classification

Abstract In motor imagery-based brain-computer interfaces (BCIs), the spatial covariance features of electroencephalography (EEG) signals that lie on Riemannian manifolds are used to enhance classification performance imagery BCIs. However, problem subject-specific bandpass frequency selection frequently arises in manifold-based methods. this study, we propose a multiple graph fusion (MRGF) model optimize band for manifold. After constructing graphs corresponding bands, embedding based bilinear mapping and mutual information were applied simultaneously extract spectral EEG from graphs. Furthermore, with support vector machine (SVM) classifier performed learned features, obtained an efficient algorithm, which achieves higher various datasets, such as BCI competition IIa in-house datasets. The proposed methods can also be other problems sample data form matrices.