Breaking Spatial Boundaries: Spectral-Domain Registration Guided Hyperspectral and Multispectral Blind Fusion

2025-06-25

AI Summary

Blind fusion of unregistered hyperspectral images (HSIs) and multispectral images (MSIs) remains challenging, particularly when large spatial resolution disparities and high-resolution remote sensing imagery render conventional spatial registration computationally prohibitive. Method: This paper innovatively shifts the registration task from the spatial to the spectral domain, proposing a spectral-prior learning network for blind HSI–MSI fusion. The framework integrates subspace representation, group sparsity regularization, and a cyclic training strategy within a lightweight network architecture, and employs a proximal alternating optimization algorithm that avoids explicit rank estimation for efficient solution. Contribution/Results: Extensive experiments on both synthetic and real remote sensing datasets demonstrate that the proposed method significantly improves registration accuracy and fusion quality. Classification performance increases by an average of 3.2%, while computational time decreases by approximately 40%, achieving a favorable balance between high accuracy and high efficiency.

Technology Category

• Computer Vision: Multi-modal Vision
• 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 blind fusion of unregistered hyperspectral images (HSIs) and multispectral images (MSIs) has attracted growing attention recently. To address the registration challenge, most existing methods employ spatial transformations on the HSI to achieve alignment with the MSI. However, due to the substantial differences in spatial resolution of the images, the performance of these methods is often unsatisfactory. Moreover, the registration process tends to be time-consuming when dealing with large-sized images in remote sensing. To address these issues, we propose tackling the registration problem from the spectral domain. Initially, a lightweight Spectral Prior Learning (SPL) network is developed to extract spectral features from the HSI and enhance the spectral resolution of the MSI. Following this, the obtained image undergoes spatial downsampling to produce the registered HSI. In this process, subspace representation and cyclic training strategy are employed to improve spectral accuracy of the registered HSI obtained. Next, we propose a blind sparse fusion (BSF) method, which utilizes group sparsity regularization to equivalently promote the low-rankness of the image. This approach not only circumvents the need for rank estimation, but also reduces computational complexity. Then, we employ the Proximal Alternating Optimization (PAO) algorithm to solve the BSF model, and present its convergence analysis. Finally, extensive numerical experiments on simulated and real datasets are conducted to verify the effectiveness of our method in registration and fusion. We also demonstrate its efficacy in enhancing classification performance.

Problem

• Fusion of unregistered hyperspectral and multispectral images
• Addressing spectral and spatial resolution differences
• Reducing computational complexity in registration and fusion

Innovation

• Spectral-domain registration via lightweight SPL network
• Blind sparse fusion with group sparsity regularization
• Proximal Alternating Optimization for efficient convergence