Dual-Domain Manifold Modeling for Hyperspectral Image Fusion

2026-07-28

AI Summary

This work addresses the challenge that existing hyperspectral image fusion methods often suffer from spatial structural degradation and spectral distortion due to their inability to effectively model geometric constraints. To overcome this limitation, the paper proposes a dual-domain manifold modeling framework that jointly introduces manifold structures in both spatial and spectral domains. A topology-aware Transformer is developed to simultaneously capture spatial topology and pixel-level manifold relationships. Furthermore, a frequency-domain decoupled fusion module is designed to separate high- and low-frequency components, thereby enhancing high-frequency geometric details and spectral reconstruction fidelity. Integrating discrete cosine transform, low-rank priors, and a spectral-driven spatial enhancement strategy, the proposed method achieves state-of-the-art performance on multiple benchmark datasets, demonstrating superior spatial fidelity and spectral accuracy compared to current approaches.

Abstract

Achieving a coherent integration of spectral richness and spatial fidelity remains a central objective in hyperspectral image fusion. However, existing hyperspectral image fusion methods struggle to effectively model geometric constraints. In the spatial domain, weak spatial-spectral interaction limits geometry-aware feature learning and suppresses high-frequency structural information, resulting in low-frequency bias and structural degradation. In the spectral domain, local manifold structures induced by spectral similarity are insufficiently exploited, limiting intrinsic pixel relationship modeling and fine-grained spectral reconstruction. To address these challenges, we propose a dual-domain manifold modeling (DDMM) framework. Specifically, we introduce a Topology-Aware Transformer (TPFormer) that combines global attention with neighborhood propagation, jointly modeling spatial topology and pixel-level feature manifold relationships to capture intrinsic spatial-spectral structures and improve topology-aware representation learning. Furthermore, a Frequency-Decoupled Spatial-Spectral Collaborative Fusion (FDSCF) module is devised, in which features are projected into the frequency domain via the discrete cosine transform and explicitly decoupled into low- and high-frequency components. Guided by a low-rank structural prior and spectral-driven spatial enhancement, FDSCF selectively enhances geometry-aware high-frequency features, strengthening spatia-spectral coupling and recovering sharper edges and finer textures. Extensive experiments on multiple benchmark datasets demonstrate that DDMM achieves superior overall performance over SoTA methods in terms of spatial structure preservation and spectral reconstruction.

Problem

• hyperspectral image fusion
• geometric constraints
• spatial-spectral interaction
• manifold modeling
• structural degradation

Innovation

• Dual-Domain Manifold Modeling
• Topology-Aware Transformer
• Frequency-Decoupled Fusion
• Hyperspectral Image Fusion
• Spatial-Spectral Collaboration