Self-attention-based expression recognition method and system for stroke patient

The invention discloses a stroke patient expression recognition method and system based on self-attention, and the method comprises the steps: obtaining a stroke patient expression data set, constructing a ViT-based expression recognition model, training the ViT-based expression recognition model through a face expression data set and the stroke patient expression data set, and obtaining a stroke patient expression recognition model. And identifying the type of the expression of the stroke patient through the trained ViT-based expression identification model. According to the method, the expression recognition model based on the ViT is constructed, the facial expression data set is used as the input for pre-training to obtain the pre-training model, and then the data set obtained after the expression data set of the stroke patient is enhanced is used as the input for training the pre-training model, so that the expression recognition model based on the ViT can recognize the expression of the lightweight strok