FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification

Convolutional Neural Network- (CNN-) based land cover classification algorithms have recently been applied in hyperspectral images (HSI) field. However, the large-scale training parameters bring huge computation burden to CNN and the spatial variability of spectral signatures leads to relative low c...

Full description

Bibliographic Details
Main Authors:	Feifei Hou, Wentai Lei, Hong Li, Jingchun Xi
Format:	Article
Language:	English
Published:	Hindawi Limited 2018-01-01
Series:	Mathematical Problems in Engineering
Online Access:	http://dx.doi.org/10.1155/2018/9218092

id	doaj-875a2ebd418647108d74e72ddcb4e515
record_format	Article
spelling	doaj-875a2ebd418647108d74e72ddcb4e5152020-11-24T20:50:48ZengHindawi LimitedMathematical Problems in Engineering1024-123X1563-51472018-01-01201810.1155/2018/92180929218092FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image ClassificationFeifei Hou0Wentai Lei1Hong Li2Jingchun Xi3School of Information Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Information Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Information Science and Engineering, Central South University, Changsha 410083, ChinaSchool of Information Science and Engineering, Central South University, Changsha 410083, ChinaConvolutional Neural Network- (CNN-) based land cover classification algorithms have recently been applied in hyperspectral images (HSI) field. However, the large-scale training parameters bring huge computation burden to CNN and the spatial variability of spectral signatures leads to relative low classification accuracy. In this paper, we propose a CNN-based classification framework that extracts square matrix representation-based spectral-spatial features and performs land cover classification. Numerical results on popular datasets show that our framework outperforms sparsity-based approaches like basic thresholding classifier-weighted least squares (BTC-WLS) and other deep learning-based methods in terms of both classification accuracy and computational cost.http://dx.doi.org/10.1155/2018/9218092
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Feifei Hou Wentai Lei Hong Li Jingchun Xi
spellingShingle	Feifei Hou Wentai Lei Hong Li Jingchun Xi FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification Mathematical Problems in Engineering
author_facet	Feifei Hou Wentai Lei Hong Li Jingchun Xi
author_sort	Feifei Hou
title	FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification
title_short	FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification
title_full	FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification
title_fullStr	FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification
title_full_unstemmed	FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification
title_sort	fmrss net: fast matrix representation-based spectral-spatial feature learning convolutional neural network for hyperspectral image classification
publisher	Hindawi Limited
series	Mathematical Problems in Engineering
issn	1024-123X 1563-5147
publishDate	2018-01-01
description	Convolutional Neural Network- (CNN-) based land cover classification algorithms have recently been applied in hyperspectral images (HSI) field. However, the large-scale training parameters bring huge computation burden to CNN and the spatial variability of spectral signatures leads to relative low classification accuracy. In this paper, we propose a CNN-based classification framework that extracts square matrix representation-based spectral-spatial features and performs land cover classification. Numerical results on popular datasets show that our framework outperforms sparsity-based approaches like basic thresholding classifier-weighted least squares (BTC-WLS) and other deep learning-based methods in terms of both classification accuracy and computational cost.
url	http://dx.doi.org/10.1155/2018/9218092
work_keys_str_mv	AT feifeihou fmrssnetfastmatrixrepresentationbasedspectralspatialfeaturelearningconvolutionalneuralnetworkforhyperspectralimageclassification AT wentailei fmrssnetfastmatrixrepresentationbasedspectralspatialfeaturelearningconvolutionalneuralnetworkforhyperspectralimageclassification AT hongli fmrssnetfastmatrixrepresentationbasedspectralspatialfeaturelearningconvolutionalneuralnetworkforhyperspectralimageclassification AT jingchunxi fmrssnetfastmatrixrepresentationbasedspectralspatialfeaturelearningconvolutionalneuralnetworkforhyperspectralimageclassification
_version_	1716803583477809152

FMRSS Net: Fast Matrix Representation-Based Spectral-Spatial Feature Learning Convolutional Neural Network for Hyperspectral Image Classification

Similar Items