Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network

The hyperspectral image is a three-dimensional (3D) hypercube with spectral and spatial continuity. Traditional hyperspectral imaging (HSI) processing mainly focuses on spectral information. However, this paper proposed a new hybrid convolutional neural network (New-Hybrid-CNN) algorithm using HSI s...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Qingshuang Mu, Zhilong Kang, Yanju Guo, Lei Chen, Shenyi Wang, Yuchen Zhao
Formato:	article
Lenguaje:	EN
Publicado:	Taylor & Francis Group 2021
Materias:	hyperspectral image three-dimensional convolutional neural network depth separable convolution support vector machine classification of wolfberry origin Nutrition. Foods and food supply TX341-641 Food processing and manufacture TP368-456
Acceso en línea:	https://doaj.org/article/a7f8f9e11ffc4fdc8aa0dcc34806a2bc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a7f8f9e11ffc4fdc8aa0dcc34806a2bc
record_format	dspace
spelling	oai:doaj.org-article:a7f8f9e11ffc4fdc8aa0dcc34806a2bc2021-11-04T15:00:41ZHyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network1094-29121532-238610.1080/10942912.2021.1987457https://doaj.org/article/a7f8f9e11ffc4fdc8aa0dcc34806a2bc2021-01-01T00:00:00Zhttp://dx.doi.org/10.1080/10942912.2021.1987457https://doaj.org/toc/1094-2912https://doaj.org/toc/1532-2386The hyperspectral image is a three-dimensional (3D) hypercube with spectral and spatial continuity. Traditional hyperspectral imaging (HSI) processing mainly focuses on spectral information. However, this paper proposed a new hybrid convolutional neural network (New-Hybrid-CNN) algorithm using HSI spectral-spatial joint information. We used the algorithm combined with HSI processing to classify the origin of Chinese wolfberry from Ningxia, Qinghai, Gansu, and Xinjiang. (1) Selecting the region of interest (ROI) over the raw HSI data as input; (2) Extracting spectral-spatial joint information from the hyperspectral stack information using homogeneous 3D convolution architecture with $$3 \times 3 \times 3$$ convolution kernels; (3) Then the depth separable convolution (DSC) was used to learn spatial information. This algorithm combined the advantages of 3D convolution and DSC, and it effectively extracted deep spectral-spatial joint information and made the architecture more lightweight. 3D convolutional neural network (3D-CNN), hybrid spectral convolutional neural network (HybridSN), and support vector machine (SVM) were established to compare with the proposed method. The proposed algorithm made full use of the HSI information while reducing the number of parameters and training time involved in the network, and improved the classification accuracy. The classification accuracy of wolfberry origin reached more than 99%. Therefore, the New-Hybrid-CNN classifier combined with HSI had the potential to classify wolfberry origin and food detection.Qingshuang MuZhilong KangYanju GuoLei ChenShenyi WangYuchen ZhaoTaylor & Francis Grouparticlehyperspectral imagethree-dimensional convolutional neural networkdepth separable convolutionsupport vector machineclassification of wolfberry originNutrition. Foods and food supplyTX341-641Food processing and manufactureTP368-456ENInternational Journal of Food Properties, Vol 24, Iss 1, Pp 1705-1721 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	hyperspectral image three-dimensional convolutional neural network depth separable convolution support vector machine classification of wolfberry origin Nutrition. Foods and food supply TX341-641 Food processing and manufacture TP368-456
spellingShingle	hyperspectral image three-dimensional convolutional neural network depth separable convolution support vector machine classification of wolfberry origin Nutrition. Foods and food supply TX341-641 Food processing and manufacture TP368-456 Qingshuang Mu Zhilong Kang Yanju Guo Lei Chen Shenyi Wang Yuchen Zhao Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
description	The hyperspectral image is a three-dimensional (3D) hypercube with spectral and spatial continuity. Traditional hyperspectral imaging (HSI) processing mainly focuses on spectral information. However, this paper proposed a new hybrid convolutional neural network (New-Hybrid-CNN) algorithm using HSI spectral-spatial joint information. We used the algorithm combined with HSI processing to classify the origin of Chinese wolfberry from Ningxia, Qinghai, Gansu, and Xinjiang. (1) Selecting the region of interest (ROI) over the raw HSI data as input; (2) Extracting spectral-spatial joint information from the hyperspectral stack information using homogeneous 3D convolution architecture with $$3 \times 3 \times 3$$ convolution kernels; (3) Then the depth separable convolution (DSC) was used to learn spatial information. This algorithm combined the advantages of 3D convolution and DSC, and it effectively extracted deep spectral-spatial joint information and made the architecture more lightweight. 3D convolutional neural network (3D-CNN), hybrid spectral convolutional neural network (HybridSN), and support vector machine (SVM) were established to compare with the proposed method. The proposed algorithm made full use of the HSI information while reducing the number of parameters and training time involved in the network, and improved the classification accuracy. The classification accuracy of wolfberry origin reached more than 99%. Therefore, the New-Hybrid-CNN classifier combined with HSI had the potential to classify wolfberry origin and food detection.
format	article
author	Qingshuang Mu Zhilong Kang Yanju Guo Lei Chen Shenyi Wang Yuchen Zhao
author_facet	Qingshuang Mu Zhilong Kang Yanju Guo Lei Chen Shenyi Wang Yuchen Zhao
author_sort	Qingshuang Mu
title	Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
title_short	Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
title_full	Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
title_fullStr	Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
title_full_unstemmed	Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
title_sort	hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network
publisher	Taylor & Francis Group
publishDate	2021
url	https://doaj.org/article/a7f8f9e11ffc4fdc8aa0dcc34806a2bc
work_keys_str_mv	AT qingshuangmu hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork AT zhilongkang hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork AT yanjuguo hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork AT leichen hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork AT shenyiwang hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork AT yuchenzhao hyperspectralimageclassificationofwolfberrywithdifferentgeographicaloriginsbasedonthreedimensionalconvolutionalneuralnetwork
_version_	1718444790872276992

Hyperspectral image classification of wolfberry with different geographical origins based on three-dimensional convolutional neural network

Ejemplares similares