Graph convolutional network method for small sample classification of hyperspectral images

Existing based on convolutional neural network classification method of hyperspectral images usually rules of the square area of image convolution, not widely adapt to different terrain local area distribution and geometry appearance of the image, therefore, under the condition of small sample class...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	ZUO Xibing, LIU Bing, YU Xuchu, ZHANG Pengqiang, GAO Kuiliang, ZHU Enze
Formato:	article
Lenguaje:	ZH
Publicado:	Surveying and Mapping Press 2021
Materias:	hyperspectral image classification local binary patterns graph convolutional network small sample Mathematical geography. Cartography GA1-1776
Acceso en línea:	https://doaj.org/article/79798da31a874e69a512c08e42292b43
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:79798da31a874e69a512c08e42292b43
record_format	dspace
spelling	oai:doaj.org-article:79798da31a874e69a512c08e42292b432021-11-12T02:25:59ZGraph convolutional network method for small sample classification of hyperspectral images1001-159510.11947/j.AGCS.2021.20200155https://doaj.org/article/79798da31a874e69a512c08e42292b432021-10-01T00:00:00Zhttp://xb.sinomaps.com/article/2021/1001-1595/2021-10-1358.htmhttps://doaj.org/toc/1001-1595Existing based on convolutional neural network classification method of hyperspectral images usually rules of the square area of image convolution, not widely adapt to different terrain local area distribution and geometry appearance of the image, therefore, under the condition of small sample classification performance is poorer, and figure convolution can network topology information on the map represent irregular image area of the convolution. Therefore, a hyperspectral image classification method based on graph convolution network is proposed. In this method, the spatial spectral information of the image is considered in the process of constructing the graph, and the feature information of the neighbor node is aggregated by the graph convolution network. Experimental results on three data sets, Pavia university, Indian Pines and Salinas, show that this method can achieve a high classification accuracy with a small number of training samples.ZUO XibingLIU BingYU XuchuZHANG PengqiangGAO KuiliangZHU EnzeSurveying and Mapping Pressarticlehyperspectral image classificationlocal binary patternsgraph convolutional networksmall sampleMathematical geography. CartographyGA1-1776ZHActa Geodaetica et Cartographica Sinica, Vol 50, Iss 10, Pp 1358-1369 (2021)
institution	DOAJ
collection	DOAJ
language	ZH
topic	hyperspectral image classification local binary patterns graph convolutional network small sample Mathematical geography. Cartography GA1-1776
spellingShingle	hyperspectral image classification local binary patterns graph convolutional network small sample Mathematical geography. Cartography GA1-1776 ZUO Xibing LIU Bing YU Xuchu ZHANG Pengqiang GAO Kuiliang ZHU Enze Graph convolutional network method for small sample classification of hyperspectral images
description	Existing based on convolutional neural network classification method of hyperspectral images usually rules of the square area of image convolution, not widely adapt to different terrain local area distribution and geometry appearance of the image, therefore, under the condition of small sample classification performance is poorer, and figure convolution can network topology information on the map represent irregular image area of the convolution. Therefore, a hyperspectral image classification method based on graph convolution network is proposed. In this method, the spatial spectral information of the image is considered in the process of constructing the graph, and the feature information of the neighbor node is aggregated by the graph convolution network. Experimental results on three data sets, Pavia university, Indian Pines and Salinas, show that this method can achieve a high classification accuracy with a small number of training samples.
format	article
author	ZUO Xibing LIU Bing YU Xuchu ZHANG Pengqiang GAO Kuiliang ZHU Enze
author_facet	ZUO Xibing LIU Bing YU Xuchu ZHANG Pengqiang GAO Kuiliang ZHU Enze
author_sort	ZUO Xibing
title	Graph convolutional network method for small sample classification of hyperspectral images
title_short	Graph convolutional network method for small sample classification of hyperspectral images
title_full	Graph convolutional network method for small sample classification of hyperspectral images
title_fullStr	Graph convolutional network method for small sample classification of hyperspectral images
title_full_unstemmed	Graph convolutional network method for small sample classification of hyperspectral images
title_sort	graph convolutional network method for small sample classification of hyperspectral images
publisher	Surveying and Mapping Press
publishDate	2021
url	https://doaj.org/article/79798da31a874e69a512c08e42292b43
work_keys_str_mv	AT zuoxibing graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages AT liubing graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages AT yuxuchu graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages AT zhangpengqiang graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages AT gaokuiliang graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages AT zhuenze graphconvolutionalnetworkmethodforsmallsampleclassificationofhyperspectralimages
_version_	1718431297799454720

Graph convolutional network method for small sample classification of hyperspectral images

Ejemplares similares