August 10, 2025

Multi-scale Autoencoder Suppression Strategy for Hyperspectral Image Anomaly Detection.

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Proposes a novel approach to improve anomaly detection in hyperspectral images, highlighting spatial information and reconstruction accuracy.

Multi-scale Autoencoder Suppression Strategy targets enhanced reconstruction of background in hyperspectral anomaly detection.
The method integrates Convolution and Transformer architectures to extract multi-scale features effectively.
Self-Attention Suppression reduces the impact of anomalies during network learning, improving detection precision.
Experiments on multiple datasets show superior performance compared to traditional and deep learning anomaly detection methods.

Tu et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

August 10, 2025

View Full Paper

Ask AI

Bookmark

Authors

Proposes a novel approach to improve anomaly detection in hyperspectral images, highlighting spatial information and reconstruction accuracy.

Multi-scale Autoencoder Suppression Strategy targets enhanced reconstruction of background in hyperspectral anomaly detection.
The method integrates Convolution and Transformer architectures to extract multi-scale features effectively.
Self-Attention Suppression reduces the impact of anomalies during network learning, improving detection precision.
Experiments on multiple datasets show superior performance compared to traditional and deep learning anomaly detection methods.

Tu et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark