Hyperspectral Image Classification with Spectral and Spatial Graph Using Inductive Representation Learning Network

Abstract

Convolutional neural networks (CNN) have achieved excellent performance for the hyperspectral image (HSI) classification problem due to better extracting spectral and spatial information. However, CNN can only perform convolution calculations on Euclidean datasets. To solve this problem, recently, the graph convolutional neural network (GCN) is proposed, which can be applied to the semisupervised HSI classification problem. However, the GCN is a direct push learning method, which requires all nodes to participate in the training process to get the node embedding. This may bring great computational cost for the hyperspectral data with a large number of pixels. Therefore, in this article, we propose an inductive learning method to solve the problem. It constructs the graph by sampling and aggregating (GraphSAGE) feature from a node's local neighborhood. This could greatly reduce the space complexity. Moreover, to enhance the classification performance, we also construct the graph using spectral and spatial information (spectra-spatial GraphSAGE). Experiments on several hyperspectral image datasets show that the proposed method can achieve better classification performance compared with state-of-the-art HSI classification methods.