Decoding tag prediction method and device, model establishment method and device and electronic equipment

• Main Authors: YIN WANGUANG, LIU QUANYING
• Format: Patent
• Language: Chinese; English
• Published: 06.08.2021
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses a decoding tag prediction method, a model establishment method and device and electronic equipment, and relates to the technical field of data processing. The decoding label prediction method comprises the following steps: acquiring a target electroencephalogram signal when a human body performs action imagination; filtering the target electroencephalogram signal to obtain a filtered electroencephalogram signal; based on the filtered electroencephalogram signals and a preset prediction model, obtaining a predicted decoding label used for representing the action intention, wherein the prediction model is a partial least square regression model optimized through Riemannian manifold. According to the scheme, the electroencephalogram signals are decoded through the partial least squares regression model after Riemannian manifold optimization, so that the optimization problem of the Euclidean space is converted into the optimization problem of the Riemannian manifold space, a global optimal