Abstract: The present invention relates to the technical field of expression recognition, and in particular, to an expression recognition method and system based on multi-scale features and spatial attention. The method includes: performing feature extraction on acquired facial image data by using an HNFER neural network model to obtain an original input feature map; performing pooling and concatenation on extracted features based on a CoordAtt attention mechanism to obtain a feature map; performing deep convolution processing on the feature map to obtain an attention map, and then performing element-by-element multiplication to obtain a final feature map; and performing feature transformation and normalization on the final feature map to obtain an expression category probability and output the expression category probability.

Abstract: The present invention discloses a human-robot collaboration method based on a multi-scale graph convolutional neural network. The method includes the following steps: S1, data acquisition: acquiring a dataset of a human skeleton in human-robot collaboration scenes, and performing pre-processing to obtain pre-processed data; S2, model training: loading the pre-processed data, and obtaining a human behavior recognition network model by training a multi-scale graph convolutional neural network; S3, human behavior recognition: predicting human behaviors through a trained deep learning network model; and S4, human-robot interaction: sending predicted information to a robot system through a communication algorithm, and enabling a robot to make action plans based on the human behaviors. By the human-robot collaboration method based on a multi-scale graph convolutional neural network disclosed by the present invention, a robot can predict human behaviors and intents in real scenes and make correct interaction.