Two-Branch Deeper Graph Convolutional Network for Hyperspectral Image Classification

The graph convolutional network (GCN) has recently attracted great attention in hyperspectral image (HSI) classification due to its strong ability to aggregate information of neighborhood nodes. However, a GCN model usually suffers from the oversmoothing problem (i.e., all nodes’ representations converge to a stationary point) when the number of GCN layers is increased. In addition, GCNs always work on superpixel-level nodes to reduce the computational cost, so pixel-level features cannot be well captured. To deal with these problems, a novel two-branch deeper GCN (TBDGCN) is proposed to combine the advantages of superpixel-based GCN and pixel-based CNN, which can simultaneously extract superpixel- and pixel-level features of HSIs. In the GCN branch, a GCN module with the DropEdge technique and residual connection is designed to alleviate oversmoothing and overfitting problems, which results in a deeper network structure with more than ten layers. In the CNN branch, to capture spatial positional information and channel information, a mixed attention mechanism is constructed to extract attention-based spectral–spatial features. The features of the GCN and CNN branches are then fused for classification. Experimental results on three benchmark HSI datasets show that the classification performance of our TBDGCN is better than existing GCN models, especially in the case of a small sample size.