High-Order Coupled Fully Connected Tensor Network Decomposition for Hyperspectral Image Super-Resolution

Hyperspectral image (HSI) super-resolution addresses the problem of fusing a low-resolution HSI (LR-HSI) and a high-resolution multispectral image (HR-MSI) to produce a high-resolution HSI (HR-HSI). Tensor analysis has been proven to be an efficient method for HSI processing. However, the existing tensor-based methods of HSI super-resolution (HSI-SR) like the tensor train and tensor ring decomposition only establish an operation between adjacent two factors and are highly sensitive to the permutation of tensor modes, leading to an inadequate and inflexible representation. In this letter, we propose a novel method for HSI-SR by utilizing the specific properties of high-order tensors in fully-connected tensor network decomposition (FCTN). The proposed method first tensorizes the target HR-HSI into a high-order tensor that has multiscale spatial structures. Then, a coupled FCTN model is proposed to fuse the corresponding high-order tensors of LR-HSI and HR-MSI. Moreover, a weighted-graph regularization (WGR) is imposed on the spectral core tensors to preserve spectral information. In the proposed model, the superiorities of the FCTN lie in the outstanding capability for characterizing adequately the intrinsic correlations between any two modes of tensors and the essential invariance for transposition. Experimental results on three datasets show the effectiveness of the proposed approach as compared to other HSI-SR methods.