Spectral Difference Guided Graph Attention Autoencoder for Hyperspectral Anomaly Detection

Abstract

Hyperspectral anomaly detection (HAD) aims at distinguishing anomalies from background in an unsupervised manner. Autoencoder (AE) and its variant-based methods have achieved promising detection performance in HAD. However, most existing methods neglect to exploit the local structure information of hyperspectral images (HSIs) that reflects the underlying relationships between each pixel and its surroundings. Hence, the representation capabilities of the networks are restricted. Moreover, reconstruction of anomalies during training compels the networks to learn abnormal patterns and, thus, reduces the spectral differences between background and anomalies. To address these problems, a spectral difference guided graph attention autoencoder (SDGATA) network is proposed for HAD. Specifically, the relationships among samples are modeled by a GAT encoder, where a spectral sharpening constraint is introduced to guide the attention coefficient learning. In this way, the encoder can also represent the spectral differences between central nodes and their neighbors. Then, a learnable GAT decoder is constructed to reconstruct node attributes and obtain the node reconstruction errors for HAD. Besides, a background purification method is proposed to generate superpixel-level samples to suppress anomaly reconstruction during training. Extensive quantitative and qualitative evaluations on six real datasets and one synthetic dataset show that the proposed method achieves competitive detection performance as compared with the state-of-the-art HAD methods.