Hyperspectral Super-Resolution with Inter-Image Variability via Degradation-based Low-Rank and Residual Fusion Method

2025-11-19

AI Summary

To address cross-image spectral variability and local spatial inconsistencies arising from acquisition discrepancies in hyperspectral (HSI)–multispectral (MSI) image fusion, this paper proposes a degradation modeling and low-rank residual decomposition framework. The method explicitly models spectral variability as dynamic changes in a degradation operator, while decomposing the fused representation into a low-rank component capturing global structural coherence and a residual component recovering high-frequency spatial details. Leveraging implicit regularization and a plug-and-play (PnP) strategy, it integrates an external deep denoiser to strengthen prior guidance. Efficient optimization is achieved via dimensionality reduction preprocessing and proximal alternating minimization. Experiments demonstrate robust convergence under strong cross-image spectral variability and significantly superior reconstruction accuracy compared to state-of-the-art methods.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Multi-instance/Multi-view Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The fusion of hyperspectral image (HSI) with multispectral image (MSI) provides an effective way to enhance the spatial resolution of HSI. However, due to different acquisition conditions, there may exist spectral variability and spatially localized changes between HSI and MSI, referred to as inter-image variability, which can significantly affect the fusion performance. Existing methods typically handle inter-image variability by applying direct transformations to the images themselves, which can exacerbate the ill-posedness of the fusion model. To address this challenge, we propose a Degradation-based Low-Rank and Residual Fusion (DLRRF) model. First, we model the spectral variability as change in the spectral degradation operator. Second, to recover the lost spatial details caused by spatially localized changes, we decompose the target HSI into low rank and residual components, where the latter is used to capture the lost details. By exploiting the spectral correlation within the images, we perform dimensionality reduction on both components. Additionally, we introduce an implicit regularizer to utilize the spatial prior information from the images. The proposed DLRRF model is solved using the Proximal Alternating Optimization (PAO) algorithm within a Plug-and-Play (PnP) framework, where the subproblem regarding implicit regularizer is addressed by an external denoiser. We further provide a comprehensive convergence analysis of the algorithm. Finally, extensive numerical experiments demonstrate that DLRRF achieves superior performance in fusing HSI and MSI with inter-image variability.

Problem

• Addresses spectral variability and spatial changes in HSI-MSI fusion
• Models spectral variability as degradation operator changes
• Recovers lost spatial details via low-rank residual decomposition

Innovation

• Models spectral variability via degradation operator changes
• Decomposes HSI into low-rank and residual components
• Uses PAO algorithm with Plug-and-Play denoiser framework