Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks

Hyperspectral unmixing refers to the process of obtaining endmembers and abundance vectors through linear or nonlinear models. The traditional linear unmixing model assumes that each mixed pixel can be represented by a linear combination of endmembers. Considering real-world situations, a sparse con...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Lulu Wan, Tao Chen, Antonio Plaza, Haojie Cai
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Convolutional neural networks (CNNs) hyperspectral unmixing spectral information Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
Acceso en línea:	https://doaj.org/article/3ab52ff928214e38a290134dfd635142
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:3ab52ff928214e38a290134dfd635142
record_format	dspace
spelling	oai:doaj.org-article:3ab52ff928214e38a290134dfd6351422021-12-01T00:00:21ZHyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks2151-153510.1109/JSTARS.2021.3126755https://doaj.org/article/3ab52ff928214e38a290134dfd6351422021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9609631/https://doaj.org/toc/2151-1535Hyperspectral unmixing refers to the process of obtaining endmembers and abundance vectors through linear or nonlinear models. The traditional linear unmixing model assumes that each mixed pixel can be represented by a linear combination of endmembers. Considering real-world situations, a sparse constraint is normally added to the linear unmixing model. However, the linear model does not take into account that the spectrum of mixed pixels is not simply linearly mixed. To fully study the mixing characteristics of ground object spectra before being imaged by the sensor, we propose a supervised unmixing architecture based on a one-dimensional convolutional neural network (CNN) by considering the spectral information and the sparse characteristics in the mixed pixel. Since 1-D CNN only considers feature learning, we combine the traditional root-mean-square error (RMSE) and <inline-formula><tex-math notation="LaTeX">${\ell _{1}}$</tex-math></inline-formula> regularization in its loss function to minimize training error. The performance of our proposed unmixing model is assessed by comparing the unmixing results with three traditional linear sparse unmixing algorithms and the fuzzy ARTMAP neural network in a simulated dataset and three real datasets. The RMSE was used to verify the unmixing accuracy of the different methods. The results showed that the RMSE obtained by our proposed CNN-based method was the lowest among the methods on all three real datasets, proving the effectiveness and stability of the CNN in unmixing tasks.Lulu WanTao ChenAntonio PlazaHaojie CaiIEEEarticleConvolutional neural networks (CNNs)hyperspectral unmixingspectral informationOcean engineeringTC1501-1800Geophysics. Cosmic physicsQC801-809ENIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 14, Pp 11669-11682 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Convolutional neural networks (CNNs) hyperspectral unmixing spectral information Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809
spellingShingle	Convolutional neural networks (CNNs) hyperspectral unmixing spectral information Ocean engineering TC1501-1800 Geophysics. Cosmic physics QC801-809 Lulu Wan Tao Chen Antonio Plaza Haojie Cai Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
description	Hyperspectral unmixing refers to the process of obtaining endmembers and abundance vectors through linear or nonlinear models. The traditional linear unmixing model assumes that each mixed pixel can be represented by a linear combination of endmembers. Considering real-world situations, a sparse constraint is normally added to the linear unmixing model. However, the linear model does not take into account that the spectrum of mixed pixels is not simply linearly mixed. To fully study the mixing characteristics of ground object spectra before being imaged by the sensor, we propose a supervised unmixing architecture based on a one-dimensional convolutional neural network (CNN) by considering the spectral information and the sparse characteristics in the mixed pixel. Since 1-D CNN only considers feature learning, we combine the traditional root-mean-square error (RMSE) and <inline-formula><tex-math notation="LaTeX">${\ell _{1}}$</tex-math></inline-formula> regularization in its loss function to minimize training error. The performance of our proposed unmixing model is assessed by comparing the unmixing results with three traditional linear sparse unmixing algorithms and the fuzzy ARTMAP neural network in a simulated dataset and three real datasets. The RMSE was used to verify the unmixing accuracy of the different methods. The results showed that the RMSE obtained by our proposed CNN-based method was the lowest among the methods on all three real datasets, proving the effectiveness and stability of the CNN in unmixing tasks.
format	article
author	Lulu Wan Tao Chen Antonio Plaza Haojie Cai
author_facet	Lulu Wan Tao Chen Antonio Plaza Haojie Cai
author_sort	Lulu Wan
title	Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
title_short	Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
title_full	Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
title_fullStr	Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
title_full_unstemmed	Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks
title_sort	hyperspectral unmixing based on spectral and sparse deep convolutional neural networks
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/3ab52ff928214e38a290134dfd635142
work_keys_str_mv	AT luluwan hyperspectralunmixingbasedonspectralandsparsedeepconvolutionalneuralnetworks AT taochen hyperspectralunmixingbasedonspectralandsparsedeepconvolutionalneuralnetworks AT antonioplaza hyperspectralunmixingbasedonspectralandsparsedeepconvolutionalneuralnetworks AT haojiecai hyperspectralunmixingbasedonspectralandsparsedeepconvolutionalneuralnetworks
_version_	1718406195973193728

Hyperspectral Unmixing Based on Spectral and Sparse Deep Convolutional Neural Networks

Ejemplares similares