Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature

Hyperspectral (HS) remote sensing imagery has broad applications but faces challenges in acquiring high-quality wide-swath data due to sensor hardware limitations. To address this, we proposed a jointly fused convolutional neural network for spectral super-resolution (JF-CNNSSR), leveraging spectral...

Full description

Saved in:
Bibliographic Details
Main Authors: Wang Benlin, Yu Qinglin, Wang Zuo, Li Weitao, Wang Yong, Liu Huan, Gu Shuangxi, Zhang Lingling, Lv Dong
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/11045064/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850117365185904640
author Wang Benlin
Yu Qinglin
Wang Zuo
Li Weitao
Wang Yong
Liu Huan
Gu Shuangxi
Zhang Lingling
Lv Dong
author_facet Wang Benlin
Yu Qinglin
Wang Zuo
Li Weitao
Wang Yong
Liu Huan
Gu Shuangxi
Zhang Lingling
Lv Dong
author_sort Wang Benlin
collection DOAJ
description Hyperspectral (HS) remote sensing imagery has broad applications but faces challenges in acquiring high-quality wide-swath data due to sensor hardware limitations. To address this, we proposed a jointly fused convolutional neural network for spectral super-resolution (JF-CNNSSR), leveraging spectral reflectance variations across different land cover types. The method employs paired overlapping multispectral (MS) and HS images to develop two spectral mapping networks: a 1-D-CNN spectral mapping network and a 3-D-CNN spectral mapping network, trained class-specifically on clustered spectral samples. A spectral distance-based joint weighting mechanism integrates outputs from both networks to synthesize HS imagery spatially consistent with the input MS data. Experiments on four benchmark datasets demonstrate that JF-CNNSSR outperforms existing spectral super-resolution methods, achieving significant improvements across key quantitative metrics including peak signal-to-noise ratio, spectral angle mapper, structural similarity index measure, ERGAS, and MSE. Practical validation through grassland coverage estimation in China’s Three-River Source Region further confirms the superiority of reconstructed HS imagery over original MS data in ecological application accuracy.
format Article
id doaj-art-e35757a0cd234086a12448abacfdec0f
institution OA Journals
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-e35757a0cd234086a12448abacfdec0f2025-08-20T02:36:06ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118162271624510.1109/JSTARS.2025.358133811045064Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral FeatureWang Benlin0https://orcid.org/0000-0001-6677-3813Yu Qinglin1https://orcid.org/0009-0008-9257-4832Wang Zuo2Li Weitao3Wang Yong4Liu Huan5Gu Shuangxi6Zhang Lingling7Lv Dong8School of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaAnhui Normal University, Wuhu, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaSchool of Geographic Information and Tourism, Chuzhou University, Chuzhou, ChinaHyperspectral (HS) remote sensing imagery has broad applications but faces challenges in acquiring high-quality wide-swath data due to sensor hardware limitations. To address this, we proposed a jointly fused convolutional neural network for spectral super-resolution (JF-CNNSSR), leveraging spectral reflectance variations across different land cover types. The method employs paired overlapping multispectral (MS) and HS images to develop two spectral mapping networks: a 1-D-CNN spectral mapping network and a 3-D-CNN spectral mapping network, trained class-specifically on clustered spectral samples. A spectral distance-based joint weighting mechanism integrates outputs from both networks to synthesize HS imagery spatially consistent with the input MS data. Experiments on four benchmark datasets demonstrate that JF-CNNSSR outperforms existing spectral super-resolution methods, achieving significant improvements across key quantitative metrics including peak signal-to-noise ratio, spectral angle mapper, structural similarity index measure, ERGAS, and MSE. Practical validation through grassland coverage estimation in China’s Three-River Source Region further confirms the superiority of reconstructed HS imagery over original MS data in ecological application accuracy.https://ieeexplore.ieee.org/document/11045064/Deep learninghyperspectral imaging (HSI)image fusionmultispectral imaging (MSI)spectral super-resolution (SSR)
spellingShingle Wang Benlin
Yu Qinglin
Wang Zuo
Li Weitao
Wang Yong
Liu Huan
Gu Shuangxi
Zhang Lingling
Lv Dong
Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep learning
hyperspectral imaging (HSI)
image fusion
multispectral imaging (MSI)
spectral super-resolution (SSR)
title Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
title_full Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
title_fullStr Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
title_full_unstemmed Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
title_short Spectral Super-Resolution Reconstruction of Multispectral Remote Sensing Images via Clustering-Based Spectral Feature
title_sort spectral super resolution reconstruction of multispectral remote sensing images via clustering based spectral feature
topic Deep learning
hyperspectral imaging (HSI)
image fusion
multispectral imaging (MSI)
spectral super-resolution (SSR)
url https://ieeexplore.ieee.org/document/11045064/
work_keys_str_mv AT wangbenlin spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT yuqinglin spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT wangzuo spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT liweitao spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT wangyong spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT liuhuan spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT gushuangxi spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT zhanglingling spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature
AT lvdong spectralsuperresolutionreconstructionofmultispectralremotesensingimagesviaclusteringbasedspectralfeature