RSAAE: Residual Self-Attention-Based Autoencoder for Hyperspectral Anomaly Detection

Autoencoder (AE) has been widely used in the field of hyperspectral anomaly detection. It is assumed that the background can be reconstructed well, but the anomalies cannot. Hence, the pixels with larger reconstruction errors are considered anomalies. However, owing to the strong nonlinear represent...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing 2023, Vol.61, p.1-14
Main Authors: Wang, Liguo, Wang, Xiaoyi, Vizziello, Anna, Gamba, Paolo
Format: Article
Language:English
Subjects:
Anomalies
Anomaly detection
Autoencoder (AE)
Coders
Decoding
Detection
Dictionaries
Feature extraction
hyperspectral anomaly detection
Hyperspectral imaging
Image reconstruction
low-rank loss function
Manifolds (mathematics)
Modules
reconstruction error
residual self-attention (RSA) module
Sparse matrices
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Autoencoder (AE) has been widely used in the field of hyperspectral anomaly detection. It is assumed that the background can be reconstructed well, but the anomalies cannot. Hence, the pixels with larger reconstruction errors are considered anomalies. However, owing to the strong nonlinear representation ability of AE, it is difficult to distinguish between background and anomalies. To address this problem, we propose a residual self-attention-based AE (RSAAE) for hyperspectral anomaly detection. RSAAE consists of dense residual self-attention (RSA) modules, an encoder, and a decoder. First, a novel RSA module is designed, which can effectively extract the main features and weaken the ability of subsequent networks to reconstruct anomalies, as well as preserve the original features to avoid the deterioration of network performance after the use of dense self-attention modules. Furthermore, inspired by manifold learning, we assume that the background is low-rank in the original space, and has the same property in the latent space after dimensionality reduction. We proposed a low-rank loss function to constrain the latent space, thereby suppressing anomaly reconstruction. Experiments on four real hyperspectral image (HSI) datasets showed that the proposed RSAAE method can produce more accurate detection results than eight popular methods.
ISSN:0196-2892
1558-0644
DOI:10.1109/TGRS.2023.3271719