A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution

Hyperspectral image super-resolution (HSI SR) has achieved remarkable success with deep neural networks. Currently, most methods in HSI SR assume a predetermined degradation model during training to synthesize low-resolution images. These methods falter when confronted with HSI exhibiting degradatio...

Full description

Saved in:
Bibliographic Details
Main Authors: Shuying Li, Ruichao Sun, San Zhang, Qiang Li
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/10891577/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849393886335598592
author Shuying Li
Ruichao Sun
San Zhang
Qiang Li
author_facet Shuying Li
Ruichao Sun
San Zhang
Qiang Li
author_sort Shuying Li
collection DOAJ
description Hyperspectral image super-resolution (HSI SR) has achieved remarkable success with deep neural networks. Currently, most methods in HSI SR assume a predetermined degradation model during training to synthesize low-resolution images. These methods falter when confronted with HSI exhibiting degradation patterns and their limited flexibility restricts practical application. In addition, these methods focus on the complex network designs for superior performance, which entail high resource consumption and limit their broad application. To address these issues, in this article, we propose a dual-strategy learning framework exploring meta-transfer learning for HSI blind SR. This framework can be applied to any SR network and facilitate performance enhancement. First, we pretrain a three-channel SR model on natural image data to address the issue of insufficient HSI data. Furthermore, we innovatively propose a transfer scheme, which directly applies our pretrained three-channel SR model to HSI, thereby significantly enhancing the spectral fidelity. To enhance the model's performance under specific degradation conditions, we incorporate meta-learning, enabling it to adapt to input images after a few iterations. Besides, we introduce attention-based knowledge distillation to equip our final network with the implicit representation capability of a meta network under a lightweight premise. Extensive experiments on three benchmark datasets demonstrate that the proposed approach outperforms existing methods in various degradations.
format Article
id doaj-art-9f1955f5817c4dc69048cb0159fa727a
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-9f1955f5817c4dc69048cb0159fa727a2025-08-20T03:40:15ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01187480749410.1109/JSTARS.2025.354276610891577A Dual-Strategy Learning Framework for Hyperspectral Image Super-ResolutionShuying Li0https://orcid.org/0000-0002-9641-3584Ruichao Sun1https://orcid.org/0009-0007-6026-1033San Zhang2Qiang Li3https://orcid.org/0000-0002-6736-3389School of Automation, Xi'an University of Posts and Telecommunications, Xi'an, ChinaSchool of Automation, Xi'an University of Posts and Telecommunications, Xi'an, ChinaSchool of Automation, Xi'an University of Posts and Telecommunications, Xi'an, ChinaSchool of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi'an, ChinaHyperspectral image super-resolution (HSI SR) has achieved remarkable success with deep neural networks. Currently, most methods in HSI SR assume a predetermined degradation model during training to synthesize low-resolution images. These methods falter when confronted with HSI exhibiting degradation patterns and their limited flexibility restricts practical application. In addition, these methods focus on the complex network designs for superior performance, which entail high resource consumption and limit their broad application. To address these issues, in this article, we propose a dual-strategy learning framework exploring meta-transfer learning for HSI blind SR. This framework can be applied to any SR network and facilitate performance enhancement. First, we pretrain a three-channel SR model on natural image data to address the issue of insufficient HSI data. Furthermore, we innovatively propose a transfer scheme, which directly applies our pretrained three-channel SR model to HSI, thereby significantly enhancing the spectral fidelity. To enhance the model's performance under specific degradation conditions, we incorporate meta-learning, enabling it to adapt to input images after a few iterations. Besides, we introduce attention-based knowledge distillation to equip our final network with the implicit representation capability of a meta network under a lightweight premise. Extensive experiments on three benchmark datasets demonstrate that the proposed approach outperforms existing methods in various degradations.https://ieeexplore.ieee.org/document/10891577/Blind super-resolution (SR)hyperspectral image (HSI)knowledge distillationmeta learning
spellingShingle Shuying Li
Ruichao Sun
San Zhang
Qiang Li
A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Blind super-resolution (SR)
hyperspectral image (HSI)
knowledge distillation
meta learning
title A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
title_full A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
title_fullStr A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
title_full_unstemmed A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
title_short A Dual-Strategy Learning Framework for Hyperspectral Image Super-Resolution
title_sort dual strategy learning framework for hyperspectral image super resolution
topic Blind super-resolution (SR)
hyperspectral image (HSI)
knowledge distillation
meta learning
url https://ieeexplore.ieee.org/document/10891577/
work_keys_str_mv AT shuyingli adualstrategylearningframeworkforhyperspectralimagesuperresolution
AT ruichaosun adualstrategylearningframeworkforhyperspectralimagesuperresolution
AT sanzhang adualstrategylearningframeworkforhyperspectralimagesuperresolution
AT qiangli adualstrategylearningframeworkforhyperspectralimagesuperresolution
AT shuyingli dualstrategylearningframeworkforhyperspectralimagesuperresolution
AT ruichaosun dualstrategylearningframeworkforhyperspectralimagesuperresolution
AT sanzhang dualstrategylearningframeworkforhyperspectralimagesuperresolution
AT qiangli dualstrategylearningframeworkforhyperspectralimagesuperresolution