Bayesian Fully-Connected Tensor Network for Hyperspectral-Multispectral Image Fusion
Linsong Shan, Zecan Yang, Laurence T. Yang, Changlong Li, Honglu Zhao, Xin Nie
TL;DR
The paper addresses the challenge of fusing hyperspectral and multispectral images without distorting spatial-spectral structures or relying on extensive manual parameter tuning. It introduces BFCTN, a Bayesian framework that integrates a Fully-Connected Tensor Network with hierarchical sparsity priors and a variational Bayesian EM learning scheme to connect factor tensors. The approach yields state-of-the-art fusion accuracy and robustness across multiple datasets (CAVE, Harvard, Pavia) and realistic noise and degradation scenarios, with demonstrated applicability to real satellite data. The work offers practical impact by providing a robust, interpretable fusion method that reduces manual tuning and handles diverse degradations, enabling reliable high-resolution hyperspectral reconstructions.
Abstract
Tensor decomposition is a powerful tool for data analysis and has been extensively employed in the field of hyperspectral-multispectral image fusion (HMF). Existing tensor decomposition-based fusion methods typically rely on disruptive data vectorization/reshaping or impose rigid constraints on the arrangement of factor tensors, hindering the preservation of spatial-spectral structures and the modeling of cross-dimensional correlations. Although recent advances utilizing the Fully-Connected Tensor Network (FCTN) decomposition have partially alleviated these limitations, the process of reorganizing data into higher-order tensors still disrupts the intrinsic spatial-spectral structure. Furthermore, these methods necessitate extensive manual parameter tuning and exhibit limited robustness against noise and spatial degradation. To alleviate these issues, we propose the Bayesian FCTN (BFCTN) method. Within this probabilistic framework, a hierarchical sparse prior that characterizing the sparsity of physical elements, establishes connections between the factor tensors. This framework explicitly models the intrinsic physical coupling among spatial structures, spectral signatures, and local scene homogeneity. For model learning, we develop a parameter estimation method based on Variational Bayesian inference (VB) and the Expectation-Maximization (EM) algorithm, which significantly reduces the need for manual parameter tuning. Extensive experiments demonstrate that BFCTN not only achieves state-of-the-art fusion accuracy and strong robustness but also exhibits practical applicability in complex real-world scenarios.