Multi-stage multi-head attention-guided heterogeneous graph network SAR target identification method and device

The invention relates to a multi-stage multi-head attention-guided heterogeneous graph network SAR target recognition method and device and computer equipment, and the method comprises the steps: constructing corresponding heterogeneous graph data according to the related parameters of a scattering center of SAR complex signal data, and inputting the data into a convolution layer comprising multiple heterogeneous graphs; a target identification neural network of a full-graph attention mechanism layer and a classification layer, and the network carries out information fusion on nodes in graph data through node-level, semantic-level and full-graph-level attention mechanisms in sequence, so that local feature representation capable of representing an input SAR image is obtained; the target recognition is further performed through the local feature representation data obtained based on the electromagnetic scattering features, so that the interpretability of the neural network is improved while the target recognit