High-resolution remote sensing image building extraction method based on superpixels and image convolution

The invention discloses a high-resolution remote sensing image building extraction method based on superpixels and image convolution, and the method comprises the following steps: obtaining a high-resolution remote sensing image and a corresponding label image, and carrying out the preprocessing of the high-resolution remote sensing image and the corresponding label image, and obtaining a preprocessing result; inputting a preprocessing result into a trained super-pixel segmentation network, and finally obtaining a super-pixel feature; node features of the graph are constructed through the superpixel features, and the graph is obtained according to the node features of the graph and edge construction of nodes; sending the constructed graph into a to-be-trained topological graph convolutional neural network for training to obtain a trained topological graph convolutional neural network; and inputting a to-be-processed graph structure into the trained graph convolutional neural network to obtain feature vectors