Cross-subject electroencephalogram emotion recognition method based on Riemannian manifold

• Main Authors: SUN XINYU, ZHANG QIZHONG, GAO YUNYUAN, SHE QINGSHAN, ZHANG QUANJUAN
• Format: Patent
• Language: Chinese; English
• Published: 19.09.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; DIAGNOSIS; ELECTRIC DIGITAL DATA PROCESSING; HUMAN NECESSITIES; HYGIENE; IDENTIFICATION; MEDICAL OR VETERINARY SCIENCE; PHYSICS; SURGERY

The invention provides a cross-subject electroencephalogram emotion recognition method based on Riemannian manifold. A current cross-subject emotion recognition model based on electroencephalogram signals has the problems of poor model generalization, low recognition accuracy and the like, analysis is limited to amplitude information of the electroencephalogram signals, and the difference between subjects cannot be effectively reduced. According to the method, firstly, an FIR filter is used for extracting four frequency bands of theta, alpha, beta and gamma of electroencephalogram signals, then PLV matrixes between electroencephalogram signal channels are calculated, and the PLV matrix in the task state is aligned to the PLV matrix in the resting state on the Riemannian manifold, so that the difference between different subjects is reduced. And finally, mapping the PLV matrix into a tangent space for classification and identification. A titer and awakening degree binary classification experiment and a titer-a