A New Subject-Specific Discriminative and Multi-Scale Filter Bank Tangent Space Mapping Method for Recognition of Multiclass Motor Imagery

• Published in: Frontiers in human neuroscience Vol. 15; p. 595723
• Main Authors: Wu, Fan, Gong, Anmin, Li, Hongyun, Zhao, Lei, Zhang, Wei, Fu, Yunfa
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 08.03.2021; Frontiers Media S.A
• Subjects: Accuracy; Algorithms; Classification; Competition; Deep learning; Discriminant analysis; discriminative and multiscale filter bank; EEG; electroencephalogram; Electroencephalography; Mapping; Mental task performance; multiclass motor-imagery; Multivariate analysis; Neural networks; Neuroscience; Riemannian geometry based classifier; Signal processing; tangent space mapping; multiclass motor-imagery; discriminative and multiscale filter bank; tangent space mapping; electroencephalogram; Riemannian geometry based classifier

Objective: Tangent Space Mapping (TSM) using the geometric structure of the covariance matrices is an effective method to recognize multiclass motor imagery (MI). Compared with the traditional CSP method, the Riemann geometric method based on TSM takes into account the nonlinear information containe...