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

TL;DR

DMS2F-HAD introduces a dual-branch Mamba-based neural network that efficiently captures spatial and spectral features for hyperspectral anomaly detection, achieving state-of-the-art accuracy and faster inference on multiple datasets.

Contribution

It presents a novel dual-branch Mamba-based architecture with dynamic fusion for hyperspectral anomaly detection, improving efficiency and accuracy over existing methods.

Findings

Achieves an average AUC of 98.78% across 14 datasets.

Increases inference speed by 4.6 times compared to similar methods.

Demonstrates strong generalization and scalability in practical applications.

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.