TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection
Change detection (CD) is a hot issue in the research of remote sensing technology. Hyperspectral images (HSIs) greatly promote the development of CD technology because of their high resolution in the spectral domain. However, some traditional CD methods currently applied to low-dimensional and multi...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9253995/ |
id |
doaj-ce48051041944536bbd08e1a48cbbbf5 |
---|---|
record_format |
Article |
spelling |
doaj-ce48051041944536bbd08e1a48cbbbf52021-06-04T23:00:12ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-011437738810.1109/JSTARS.2020.30370709253995TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change DetectionTianming Zhan0https://orcid.org/0000-0001-5030-3032Bo Song1Le Sun2https://orcid.org/0000-0001-6465-8678Xiuping Jia3https://orcid.org/0000-0001-9916-6382Minghua Wan4Guowei Yang5https://orcid.org/0000-0002-5204-1766Zebin Wu6https://orcid.org/0000-0002-7162-0202Collaborative Innovation Center of Audit Information Engineering and Technology and the School of Information Engineering, Nanjing Audit University, Nanjing, ChinaSchool of Information Engineering, Nanjing Audit University, Nanjing, ChinaSchool of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, ChinaSchool of Engineering and Information Technology, The University of New South Wales, Canberra, BC, AustraliaSchool of Information Engineering, Nanjing Audit University, Nanjing, ChinaSchool of Information Engineering, Nanjing Audit University, Nanjing, ChinaNanjing University of Science and Technology, Nanjing, ChinaChange detection (CD) is a hot issue in the research of remote sensing technology. Hyperspectral images (HSIs) greatly promote the development of CD technology because of their high resolution in the spectral domain. However, some traditional CD methods currently applied to low-dimensional and multispectral images cannot adapt to the complex high-dimensional features of the HSIs. In addition, the spectral measurements of the HSI contain a lot of noise and redundancy, which greatly contaminates spectral-only information for CD. In order to fully extract the discriminant features of HSI to improve the accuracy of CD, this article proposes a three-directions spectral-spatial convolution neural network (TDSSC). A novel method for three-direction decomposition of hyperspectral change tensors is proposed-change tensor is decomposed along the spectral direction and two spatial directions to get a single tensor containing the spectral information and two kinds of tensors containing the spectral-spatial information. TDSSC uses 1-D convolution to extract spectral features from the spectral direction as well as reducing the tensor dimension, which helps the latter network to be lightweight and significantly improves the speed of change detection. Also, it uses 2-D convolution to extract spectral-spatial features from two spatial directions of the reduced tensor, and to extract features from different directions to improve the accuracy and Kappa value of CD. The experimental results of three real hyperspectral datasets show that TDSSC is superior to most existing CD methods.https://ieeexplore.ieee.org/document/9253995/Change detection (CD)hyperspectral image (HSI)spectral–spatial combinationthree directions convolution neural network |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Tianming Zhan Bo Song Le Sun Xiuping Jia Minghua Wan Guowei Yang Zebin Wu |
spellingShingle |
Tianming Zhan Bo Song Le Sun Xiuping Jia Minghua Wan Guowei Yang Zebin Wu TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Change detection (CD) hyperspectral image (HSI) spectral–spatial combination three directions convolution neural network |
author_facet |
Tianming Zhan Bo Song Le Sun Xiuping Jia Minghua Wan Guowei Yang Zebin Wu |
author_sort |
Tianming Zhan |
title |
TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection |
title_short |
TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection |
title_full |
TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection |
title_fullStr |
TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection |
title_full_unstemmed |
TDSSC: A Three-Directions Spectral–Spatial Convolution Neural Network for Hyperspectral Image Change Detection |
title_sort |
tdssc: a three-directions spectral–spatial convolution neural network for hyperspectral image change detection |
publisher |
IEEE |
series |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
issn |
2151-1535 |
publishDate |
2021-01-01 |
description |
Change detection (CD) is a hot issue in the research of remote sensing technology. Hyperspectral images (HSIs) greatly promote the development of CD technology because of their high resolution in the spectral domain. However, some traditional CD methods currently applied to low-dimensional and multispectral images cannot adapt to the complex high-dimensional features of the HSIs. In addition, the spectral measurements of the HSI contain a lot of noise and redundancy, which greatly contaminates spectral-only information for CD. In order to fully extract the discriminant features of HSI to improve the accuracy of CD, this article proposes a three-directions spectral-spatial convolution neural network (TDSSC). A novel method for three-direction decomposition of hyperspectral change tensors is proposed-change tensor is decomposed along the spectral direction and two spatial directions to get a single tensor containing the spectral information and two kinds of tensors containing the spectral-spatial information. TDSSC uses 1-D convolution to extract spectral features from the spectral direction as well as reducing the tensor dimension, which helps the latter network to be lightweight and significantly improves the speed of change detection. Also, it uses 2-D convolution to extract spectral-spatial features from two spatial directions of the reduced tensor, and to extract features from different directions to improve the accuracy and Kappa value of CD. The experimental results of three real hyperspectral datasets show that TDSSC is superior to most existing CD methods. |
topic |
Change detection (CD) hyperspectral image (HSI) spectral–spatial combination three directions convolution neural network |
url |
https://ieeexplore.ieee.org/document/9253995/ |
work_keys_str_mv |
AT tianmingzhan tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT bosong tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT lesun tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT xiupingjia tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT minghuawan tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT guoweiyang tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection AT zebinwu tdsscathreedirectionsspectralx2013spatialconvolutionneuralnetworkforhyperspectralimagechangedetection |
_version_ |
1721397306777927680 |