Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification

Hyperspectral image (HSI) classification constitutes a crucial research direction within the domain of remote sensing. Convolutional neural networks (CNNs) and graph convolutional network (GCN) have exhibited outstanding classification performance in this field, emerging as current research focuses....

Full description

Saved in:
Bibliographic Details
Main Authors: Xiangyue Yu, Ning Li, Di Wu, Zheng Li, Zhenyuan Wu, Ximing Ma
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/11062325/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849246287650619392
author Xiangyue Yu
Ning Li
Di Wu
Zheng Li
Zhenyuan Wu
Ximing Ma
author_facet Xiangyue Yu
Ning Li
Di Wu
Zheng Li
Zhenyuan Wu
Ximing Ma
author_sort Xiangyue Yu
collection DOAJ
description Hyperspectral image (HSI) classification constitutes a crucial research direction within the domain of remote sensing. Convolutional neural networks (CNNs) and graph convolutional network (GCN) have exhibited outstanding classification performance in this field, emerging as current research focuses. Nevertheless, GCN possesses certain limitations in capturing the neighborhood features of images, while traditional 2-D CNNs are incapable of fully extracting the spatial information of HSI. To address these problems, we propose a novel architecture dubbed spatial multifeature and dual-layer multihop graph convolutional network (SMTGCN). This network is capable of concurrently extracting pixel-level spatial features and superpixel-level spectral features. Specifically, a dual-layer multihop graph convolutional network is constructed within the GCN branch, which can take the features of superpixel at different segmentation scales as network nodes to effectively capture and fuse the superpixel features in HSI. In the CNN branch, a multiscale spatial structure is constructed for feature extraction and fusion, and a hybrid attention mechanism model is proposed to enhance the feature capture ability, a multilayer pooling structure is added to retain more detailed information while suppressing excessive redundant data. Finally, the features extracted by the GCN branch and the CNN branch are fused to realize HSI classification. Experimental results conducted on four benchmark HSI datasets indicate that, in comparison with existing classification methods, SMTGCN achieves remarkable improvements in classification performance when using a small number of training samples.
format Article
id doaj-art-d3531edd9abc497e966ec0e4dfd15d0c
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-d3531edd9abc497e966ec0e4dfd15d0c2025-08-20T03:58:32ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118183911841010.1109/JSTARS.2025.358497011062325Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image ClassificationXiangyue Yu0https://orcid.org/0009-0002-1705-5465Ning Li1https://orcid.org/0009-0000-1885-1784Di Wu2Zheng Li3Zhenyuan Wu4Ximing Ma5Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, ChinaUniversity of Chinese Academy of Sciences, Beijing, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun, ChinaHyperspectral image (HSI) classification constitutes a crucial research direction within the domain of remote sensing. Convolutional neural networks (CNNs) and graph convolutional network (GCN) have exhibited outstanding classification performance in this field, emerging as current research focuses. Nevertheless, GCN possesses certain limitations in capturing the neighborhood features of images, while traditional 2-D CNNs are incapable of fully extracting the spatial information of HSI. To address these problems, we propose a novel architecture dubbed spatial multifeature and dual-layer multihop graph convolutional network (SMTGCN). This network is capable of concurrently extracting pixel-level spatial features and superpixel-level spectral features. Specifically, a dual-layer multihop graph convolutional network is constructed within the GCN branch, which can take the features of superpixel at different segmentation scales as network nodes to effectively capture and fuse the superpixel features in HSI. In the CNN branch, a multiscale spatial structure is constructed for feature extraction and fusion, and a hybrid attention mechanism model is proposed to enhance the feature capture ability, a multilayer pooling structure is added to retain more detailed information while suppressing excessive redundant data. Finally, the features extracted by the GCN branch and the CNN branch are fused to realize HSI classification. Experimental results conducted on four benchmark HSI datasets indicate that, in comparison with existing classification methods, SMTGCN achieves remarkable improvements in classification performance when using a small number of training samples.https://ieeexplore.ieee.org/document/11062325/Convolutional neural network (CNN)fusion attention mechanismgraph convolutional network (GCN)hyperspectral image (HSI) classification
spellingShingle Xiangyue Yu
Ning Li
Di Wu
Zheng Li
Zhenyuan Wu
Ximing Ma
Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Convolutional neural network (CNN)
fusion attention mechanism
graph convolutional network (GCN)
hyperspectral image (HSI) classification
title Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
title_full Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
title_fullStr Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
title_full_unstemmed Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
title_short Spatial Multifeature and Dual-Layer Multihop Graph Convolution Networks for Hyperspectral Image Classification
title_sort spatial multifeature and dual layer multihop graph convolution networks for hyperspectral image classification
topic Convolutional neural network (CNN)
fusion attention mechanism
graph convolutional network (GCN)
hyperspectral image (HSI) classification
url https://ieeexplore.ieee.org/document/11062325/
work_keys_str_mv AT xiangyueyu spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification
AT ningli spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification
AT diwu spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification
AT zhengli spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification
AT zhenyuanwu spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification
AT ximingma spatialmultifeatureandduallayermultihopgraphconvolutionnetworksforhyperspectralimageclassification