Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition

The fusion of a low-spatial-resolution hyperspectral image (LR-HSI) and a high-spatial-resolution multispectral image (HR-MSI) is an effective way to generate a high-resolution hyperspectral image (HR-HSI). In recent years, methods based on tensor ring (TR) decomposition have received widespread att...

Full description

Saved in:
Bibliographic Details
Main Authors: Jun Zhang, Mengling He, Chengzhi Deng
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11048944/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849432133044535296
author Jun Zhang
Mengling He
Chengzhi Deng
author_facet Jun Zhang
Mengling He
Chengzhi Deng
author_sort Jun Zhang
collection DOAJ
description The fusion of a low-spatial-resolution hyperspectral image (LR-HSI) and a high-spatial-resolution multispectral image (HR-MSI) is an effective way to generate a high-resolution hyperspectral image (HR-HSI). In recent years, methods based on tensor ring (TR) decomposition have received widespread attention due to their superior performance in approximating high-dimensional data. However, these methods often neglect the intrinsic low-rank property of TR factors. More importantly, even with low-rank consideration, their effectiveness remains severely limited by both the restrictive low-rank tensor definition and high sensitivity to the permutation of tensor modes, ultimately degrading their performance. To address these issues, we propose a new HSI–MSI fusion model based on the generalized logarithmic tensor nuclear norm (GLTNN) under the TR decomposition framework. Specifically, we extend the traditional LTNN based on the third pattern to any pattern and define the generalized LTNN, where the Fourier transform is conducted on arbitrary mode. This method can not only capture the correlations comprehensively for tensor modes, but also effectively avoid the influence of the permutation of tensor modes on the fusion results. In addition, we design a proximal alternating minimization algorithm to efficiently solve the proposed model. The experimental results on four public datasets show that our method outperforms existing approaches in both numerical metrics and visual quality, validating its effectiveness and superiority.
format Article
id doaj-art-8ecdf1b14cee4552bb3b07ba6c4f98b0
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-8ecdf1b14cee4552bb3b07ba6c4f98b02025-08-20T03:27:26ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118165961660810.1109/JSTARS.2025.358278211048944Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring DecompositionJun Zhang0https://orcid.org/0000-0003-3809-7023Mengling He1Chengzhi Deng2https://orcid.org/0000-0003-1605-7100College of Science & Key Laboratory of Engineering Mathematics and Advanced Computing, Jiangxi University of Water Resources and Electric Power, Nanchang, ChinaCollege of Science, Jiangxi University of Water Resources and Electric Power, Nanchang, ChinaJiangxi Province Key Laboratory of Smart Water Conservancy, Jiangxi University of Water Resources and Electric Power, Nanchang, ChinaThe fusion of a low-spatial-resolution hyperspectral image (LR-HSI) and a high-spatial-resolution multispectral image (HR-MSI) is an effective way to generate a high-resolution hyperspectral image (HR-HSI). In recent years, methods based on tensor ring (TR) decomposition have received widespread attention due to their superior performance in approximating high-dimensional data. However, these methods often neglect the intrinsic low-rank property of TR factors. More importantly, even with low-rank consideration, their effectiveness remains severely limited by both the restrictive low-rank tensor definition and high sensitivity to the permutation of tensor modes, ultimately degrading their performance. To address these issues, we propose a new HSI–MSI fusion model based on the generalized logarithmic tensor nuclear norm (GLTNN) under the TR decomposition framework. Specifically, we extend the traditional LTNN based on the third pattern to any pattern and define the generalized LTNN, where the Fourier transform is conducted on arbitrary mode. This method can not only capture the correlations comprehensively for tensor modes, but also effectively avoid the influence of the permutation of tensor modes on the fusion results. In addition, we design a proximal alternating minimization algorithm to efficiently solve the proposed model. The experimental results on four public datasets show that our method outperforms existing approaches in both numerical metrics and visual quality, validating its effectiveness and superiority.https://ieeexplore.ieee.org/document/11048944/Generalized logarithmic tensor nuclear norm (GLTNN)hyperspectral and multispectral image fusionproximal alternating minimization (PAM)tensor ring (TR) decomposition
spellingShingle Jun Zhang
Mengling He
Chengzhi Deng
Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Generalized logarithmic tensor nuclear norm (GLTNN)
hyperspectral and multispectral image fusion
proximal alternating minimization (PAM)
tensor ring (TR) decomposition
title Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
title_full Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
title_fullStr Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
title_full_unstemmed Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
title_short Generalized Logarithmic Tensor Nuclear Norm for Hyperspectral-Multispectral Image Fusion via Tensor Ring Decomposition
title_sort generalized logarithmic tensor nuclear norm for hyperspectral multispectral image fusion via tensor ring decomposition
topic Generalized logarithmic tensor nuclear norm (GLTNN)
hyperspectral and multispectral image fusion
proximal alternating minimization (PAM)
tensor ring (TR) decomposition
url https://ieeexplore.ieee.org/document/11048944/
work_keys_str_mv AT junzhang generalizedlogarithmictensornuclearnormforhyperspectralmultispectralimagefusionviatensorringdecomposition
AT menglinghe generalizedlogarithmictensornuclearnormforhyperspectralmultispectralimagefusionviatensorringdecomposition
AT chengzhideng generalizedlogarithmictensornuclearnormforhyperspectralmultispectralimagefusionviatensorringdecomposition