What question did this study set out to answer?

To improve hyperspectral anomaly detection by enhancing background modeling while suppressing anomalies.

Dual-stream background modeling network with anomaly suppression for hyperspectral anomaly detection

View Full Paper

Ask AI

Bookmark

Authors

Novel dual-stream network improves background modeling in hyperspectral anomaly detection, suggesting reduced anomalies.

To improve hyperspectral anomaly detection by enhancing background modeling while suppressing anomalies.
Developed a dual-stream background modeling network (DSBMNet) to address reconstruction challenges.
Implemented a multi-directional masked convolution module to disrupt spatial correlations of anomalies.
Created attention-driven multi-scale grouped convolution and difference-enhanced spatial-spectral attention modules for background feature optimization.
Ablation studies confirmed the effectiveness of the modules designed.
Comparison on six hyperspectral datasets revealed superior performance of DSBMNet.

Huo et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark