DMS2F-HAD: A Dual-branch Mamba-based Spatial-Spectral Fusion Network for Hyperspectral Anomaly Detection

2026-02-04

AI Summary

This work addresses the challenges of modeling long-range spectral dependencies, high computational cost, and limited generalization in hyperspectral anomaly detection by introducing Mamba—a state-space model—into this domain for the first time. The authors propose a dual-branch Mamba architecture that efficiently captures spatial and spectral features in parallel, complemented by a dynamic gated fusion mechanism to enhance anomaly localization. The method achieves end-to-end unsupervised learning with linear computational complexity. Evaluated on 14 benchmark datasets, it attains an average AUC of 98.78% and demonstrates a 4.6× faster inference speed compared to existing deep learning approaches, significantly outperforming current methods while exhibiting strong generalization and practical utility.

Technology Category

• Machine Learning: Learning with Manifolds
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Hyperspectral anomaly detection (HAD) aims to identify rare and irregular targets in high-dimensional hyperspectral images (HSIs), which are often noisy and unlabelled data. Existing deep learning methods either fail to capture long-range spectral dependencies (e.g., convolutional neural networks) or suffer from high computational cost (e.g., Transformers). To address these challenges, we propose DMS2F-HAD, a novel dual-branch Mamba-based model. Our architecture utilizes Mamba's linear-time modeling to efficiently learn distinct spatial and spectral features in specialized branches, which are then integrated by a dynamic gated fusion mechanism to enhance anomaly localization. Across fourteen benchmark HSI datasets, our proposed DMS2F-HAD not only achieves a state-of-the-art average AUC of 98.78%, but also demonstrates superior efficiency with an inference speed 4.6 times faster than comparable deep learning methods. The results highlight DMS2FHAD's strong generalization and scalability, positioning it as a strong candidate for practical HAD applications.

Problem

• Hyperspectral Anomaly Detection
• Long-range Spectral Dependencies
• Computational Cost
• Noisy Unlabelled Data

Innovation

• Mamba
• dual-branch architecture
• spatial-spectral fusion
• hyperspectral anomaly detection
• dynamic gated fusion