Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning

Hyperspectral image (HSI) clustering is a major challenge due to the redundant spectral information in HSIs. In this paper, we propose a novel deep subspace clustering method that extracts spatial–spectral features via contrastive learning. First, we construct positive and negative sample pairs thro...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xiang Hu, Teng Li, Tong Zhou, Yuanxi Peng
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	hyperspectral image clustering deep subspace clustering deep learning spectral clustering Science Q
Acceso en línea:	https://doaj.org/article/c6301634536045f89cc5770378d53257
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c6301634536045f89cc5770378d53257
record_format	dspace
spelling	oai:doaj.org-article:c6301634536045f89cc5770378d532572021-11-11T18:56:00ZDeep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning10.3390/rs132144182072-4292https://doaj.org/article/c6301634536045f89cc5770378d532572021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/21/4418https://doaj.org/toc/2072-4292Hyperspectral image (HSI) clustering is a major challenge due to the redundant spectral information in HSIs. In this paper, we propose a novel deep subspace clustering method that extracts spatial–spectral features via contrastive learning. First, we construct positive and negative sample pairs through data augmentation. Then, the data pairs are projected into feature space using a CNN model. Contrastive learning is conducted by minimizing the distances of positive pairs and maximizing those of negative pairs. Finally, based on their features, spectral clustering is employed to obtain the final result. Experimental results gained over three HSI datasets demonstrate that our proposed method is superior to other state-of-the-art methods.Xiang HuTeng LiTong ZhouYuanxi PengMDPI AGarticlehyperspectral image clusteringdeep subspace clusteringdeep learningspectral clusteringScienceQENRemote Sensing, Vol 13, Iss 4418, p 4418 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	hyperspectral image clustering deep subspace clustering deep learning spectral clustering Science Q
spellingShingle	hyperspectral image clustering deep subspace clustering deep learning spectral clustering Science Q Xiang Hu Teng Li Tong Zhou Yuanxi Peng Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
description	Hyperspectral image (HSI) clustering is a major challenge due to the redundant spectral information in HSIs. In this paper, we propose a novel deep subspace clustering method that extracts spatial–spectral features via contrastive learning. First, we construct positive and negative sample pairs through data augmentation. Then, the data pairs are projected into feature space using a CNN model. Contrastive learning is conducted by minimizing the distances of positive pairs and maximizing those of negative pairs. Finally, based on their features, spectral clustering is employed to obtain the final result. Experimental results gained over three HSI datasets demonstrate that our proposed method is superior to other state-of-the-art methods.
format	article
author	Xiang Hu Teng Li Tong Zhou Yuanxi Peng
author_facet	Xiang Hu Teng Li Tong Zhou Yuanxi Peng
author_sort	Xiang Hu
title	Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
title_short	Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
title_full	Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
title_fullStr	Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
title_full_unstemmed	Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning
title_sort	deep spatial-spectral subspace clustering for hyperspectral images based on contrastive learning
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/c6301634536045f89cc5770378d53257
work_keys_str_mv	AT xianghu deepspatialspectralsubspaceclusteringforhyperspectralimagesbasedoncontrastivelearning AT tengli deepspatialspectralsubspaceclusteringforhyperspectralimagesbasedoncontrastivelearning AT tongzhou deepspatialspectralsubspaceclusteringforhyperspectralimagesbasedoncontrastivelearning AT yuanxipeng deepspatialspectralsubspaceclusteringforhyperspectralimagesbasedoncontrastivelearning
_version_	1718431669912862720

Deep Spatial-Spectral Subspace Clustering for Hyperspectral Images Based on Contrastive Learning

Ejemplares similares