Joint Correlation Alignment-Based Graph Neural Network for Domain Adaptation of Multitemporal Hyperspectral Remote Sensing Images

In this article, we propose a novel deep domain adaptation method based on graph neural network (GNN) for multitemporal hyperspectral remote sensing images. In GNN, graphs are constructed for source and target data, respectively. Then the graphs are utilized in each hidden layer to obtain features. GNN operates on graph structure and the relations between data samples can be exploited. It aggregates features and propagate information through graph nodes. Thus, the extracted features have an improved smoothness in each spectral neighborhood which is beneficial to classification. Furthermore, the domain-wise correlation alignment (CORAL) and class-wise CORAL are jointly embedded in GNN network to achieve a joint distribution adaptation performance. By introducing the joint CORAL strategy in GNN, the extracted features can not only be aligned between domains but also have a superior discriminability in each domain. This domain adaptation network is named as joint CORAL-based graph neural network. Experiments using multitemporal Hyperion and NSF-funded center for airborne laser mapping datasets demonstrate the effectiveness of the proposed method.

Automated summary

This article proposes a novel deep domain adaptation method based on graph neural network for multitemporal hyperspectral remote sensing images. The method improves feature smoothness in each spectral neighborhood, beneficial for classification, and integrates domain-wise correlation alignment and class-wise CORAL for joint distribution adaptation performance. The experiments demonstrate the effectiveness of the proposed method using multitemporal Hyperion and NSF-funded center for airborne laser mapping datasets.