Electroencephalogram cognitive load analysis method and system based on dynamic graph convolution of neural imaging prior

The invention relates to the field of electroencephalogram signal recognition, in particular to an electroencephalogram cognitive load analysis method and system based on neural imaging prior dynamic graph convolution, and the method comprises the following steps: preprocessing original electroencephalogram data, processing labels, constructing a cross-subject data set of a subject, and constructing a cross-subject data set of the subject; and establishing an electroencephalogram cognitive load analysis model based on the neural imaging prior dynamic graph convolution to realize the analysis of the electroencephalogram cognitive load. In the electroencephalogram cognitive load analysis model, through a multi-scale space-time attention cavity convolution module, modeling importance between each electroencephalogram channel and time features, and extracting multi-scale feature information related to the cognitive load; priori initializing an adjacent matrix of the electroencephalogram electrode based on neuroim