Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification

In hyperspectral image (HSI) classification, convolutional neural networks (CNN) have been attracting increasing attention because of their ability to represent spectral-spatial features. Nevertheless, the conventional CNN models perform convolution operation on regular-grid image regions with a fix...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Minghua Zhang, Hongling Luo, Wei Song, Haibin Mei, Cheng Su
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	hyperspectral image classification deep learning graph convolutional network offset graph convolution spectral-spatial features Science Q
Acceso en línea:	https://doaj.org/article/cb492dd9d46f47d7ba984d8d0b09d803
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:cb492dd9d46f47d7ba984d8d0b09d803
record_format	dspace
spelling	oai:doaj.org-article:cb492dd9d46f47d7ba984d8d0b09d8032021-11-11T18:54:15ZSpectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification10.3390/rs132143422072-4292https://doaj.org/article/cb492dd9d46f47d7ba984d8d0b09d8032021-10-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/21/4342https://doaj.org/toc/2072-4292In hyperspectral image (HSI) classification, convolutional neural networks (CNN) have been attracting increasing attention because of their ability to represent spectral-spatial features. Nevertheless, the conventional CNN models perform convolution operation on regular-grid image regions with a fixed kernel size and as a result, they neglect the inherent relation between HSI data. In recent years, graph convolutional networks (GCN) used for data representation in a non-Euclidean space, have been successfully applied to HSI classification. However, conventional GCN methods suffer from a huge computational cost since they construct the adjacency matrix between all HSI pixels, and they ignore the local spatial context information of hyperspectral images. To alleviate these shortcomings, we propose a novel method termed spectral-spatial offset graph convolutional networks (SSOGCN). Different from the usually used GCN models that compute the adjacency matrix between all pixels, we construct an adjacency matrix only using pixels within a patch, which contains rich local spatial context information, while reducing the computation cost and memory consumption of the adjacency matrix. Moreover, to emphasize important local spatial information, an offset graph convolution module is proposed to extract more robust features and improve the classification performance. Comprehensive experiments are carried out on three representative benchmark data sets, and the experimental results effectively certify that the proposed SSOGCN method has more advantages than the recent state-of-the-art (SOTA) methods.Minghua ZhangHongling LuoWei SongHaibin MeiCheng SuMDPI AGarticlehyperspectral image classificationdeep learninggraph convolutional networkoffset graph convolutionspectral-spatial featuresScienceQENRemote Sensing, Vol 13, Iss 4342, p 4342 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	hyperspectral image classification deep learning graph convolutional network offset graph convolution spectral-spatial features Science Q
spellingShingle	hyperspectral image classification deep learning graph convolutional network offset graph convolution spectral-spatial features Science Q Minghua Zhang Hongling Luo Wei Song Haibin Mei Cheng Su Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
description	In hyperspectral image (HSI) classification, convolutional neural networks (CNN) have been attracting increasing attention because of their ability to represent spectral-spatial features. Nevertheless, the conventional CNN models perform convolution operation on regular-grid image regions with a fixed kernel size and as a result, they neglect the inherent relation between HSI data. In recent years, graph convolutional networks (GCN) used for data representation in a non-Euclidean space, have been successfully applied to HSI classification. However, conventional GCN methods suffer from a huge computational cost since they construct the adjacency matrix between all HSI pixels, and they ignore the local spatial context information of hyperspectral images. To alleviate these shortcomings, we propose a novel method termed spectral-spatial offset graph convolutional networks (SSOGCN). Different from the usually used GCN models that compute the adjacency matrix between all pixels, we construct an adjacency matrix only using pixels within a patch, which contains rich local spatial context information, while reducing the computation cost and memory consumption of the adjacency matrix. Moreover, to emphasize important local spatial information, an offset graph convolution module is proposed to extract more robust features and improve the classification performance. Comprehensive experiments are carried out on three representative benchmark data sets, and the experimental results effectively certify that the proposed SSOGCN method has more advantages than the recent state-of-the-art (SOTA) methods.
format	article
author	Minghua Zhang Hongling Luo Wei Song Haibin Mei Cheng Su
author_facet	Minghua Zhang Hongling Luo Wei Song Haibin Mei Cheng Su
author_sort	Minghua Zhang
title	Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
title_short	Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
title_full	Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
title_fullStr	Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
title_full_unstemmed	Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification
title_sort	spectral-spatial offset graph convolutional networks for hyperspectral image classification
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/cb492dd9d46f47d7ba984d8d0b09d803
work_keys_str_mv	AT minghuazhang spectralspatialoffsetgraphconvolutionalnetworksforhyperspectralimageclassification AT honglingluo spectralspatialoffsetgraphconvolutionalnetworksforhyperspectralimageclassification AT weisong spectralspatialoffsetgraphconvolutionalnetworksforhyperspectralimageclassification AT haibinmei spectralspatialoffsetgraphconvolutionalnetworksforhyperspectralimageclassification AT chengsu spectralspatialoffsetgraphconvolutionalnetworksforhyperspectralimageclassification
_version_	1718431670469656576

Spectral-Spatial Offset Graph Convolutional Networks for Hyperspectral Image Classification

Ejemplares similares