Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images
Hyperspectral image classification (HSIC) has generated considerable interests over the past years. However, one of challenging issues arising in HSIC is inconsistent classification, which is mainly caused by random training sampling (RTS) of selecting training data. This is because a different set...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9137652/ |
id |
doaj-d207e658693747a3822fcb5915c71b27 |
---|---|
record_format |
Article |
spelling |
doaj-d207e658693747a3822fcb5915c71b272021-06-03T23:01:05ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352020-01-01133986400710.1109/JSTARS.2020.30083599137652Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral ImagesChein-I Chang0https://orcid.org/0000-0002-5450-4891Kenneth Yeonkong Ma1https://orcid.org/0000-0003-3636-5311Chia-Chen Liang2Yi-Mei Kuo3Shuhan Chen4https://orcid.org/0000-0001-9996-0666Shengwei Zhong5https://orcid.org/0000-0001-8317-728XInformation and Technology College, Center for Hyperspectral Imaging in Remote Sensing, Dalian Maritime University, Dalian, ChinaDepartment of Computer Science and Electrical Engineering, Remote Sensing Signal and Image Processing Laboratory, University of Maryland, Baltimore, MD, USADepartment of Computer Science and Electrical Engineering, Remote Sensing Signal and Image Processing Laboratory, University of Maryland, Baltimore, MD, USADepartment of Computer Science and Electrical Engineering, Remote Sensing Signal and Image Processing Laboratory, University of Maryland, Baltimore, MD, USAZhejiang University, Hangzhou, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, ChinaHyperspectral image classification (HSIC) has generated considerable interests over the past years. However, one of challenging issues arising in HSIC is inconsistent classification, which is mainly caused by random training sampling (RTS) of selecting training data. This is because a different set of training samples may produce a different classification result. A general approach to addressing this problem is the so-called K-fold method which implements RTS K times and takes the average of overall accuracy with respect to standard deviation to describe a confidence level of classification performance. To deal with this issue, this article develops an iterative RTS (IRTS) method as an alternative to the K-fold method to reduce the uncertainty caused by RTS. Its idea is to add the spatial filtered classification maps to the image cube that is currently being processed via feedback loops to augment image cubes iteratively. Then, the training samples will be reselected randomly from the new augmented image cubes iteration-by-iteration. As a result, the training samples selected from each iteration will be updated by new added spatial information captured by spatial filters implemented at the iteration. The experimental results clearly demonstrate that IRTS successfully improves classification accuracy as well as reduces inconsistency in results.https://ieeexplore.ieee.org/document/9137652/Hyperspectral image classification (HSIC)iterative random training sampling (IRTS)<named-content xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" content-type="math" xlink:type="simple"> <inline-formula> <tex-math notation="LaTeX">$K$</tex-math> </inline-formula> </named-content>-fold methodrandom training sampling (RTS)spectral-spatial (SS) |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Chein-I Chang Kenneth Yeonkong Ma Chia-Chen Liang Yi-Mei Kuo Shuhan Chen Shengwei Zhong |
spellingShingle |
Chein-I Chang Kenneth Yeonkong Ma Chia-Chen Liang Yi-Mei Kuo Shuhan Chen Shengwei Zhong Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Hyperspectral image classification (HSIC) iterative random training sampling (IRTS) <named-content xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" content-type="math" xlink:type="simple"> <inline-formula> <tex-math notation="LaTeX">$K$</tex-math> </inline-formula> </named-content>-fold method random training sampling (RTS) spectral-spatial (SS) |
author_facet |
Chein-I Chang Kenneth Yeonkong Ma Chia-Chen Liang Yi-Mei Kuo Shuhan Chen Shengwei Zhong |
author_sort |
Chein-I Chang |
title |
Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images |
title_short |
Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images |
title_full |
Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images |
title_fullStr |
Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images |
title_full_unstemmed |
Iterative Random Training Sampling Spectral Spatial Classification for Hyperspectral Images |
title_sort |
iterative random training sampling spectral spatial classification for hyperspectral images |
publisher |
IEEE |
series |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
issn |
2151-1535 |
publishDate |
2020-01-01 |
description |
Hyperspectral image classification (HSIC) has generated considerable interests over the past years. However, one of challenging issues arising in HSIC is inconsistent classification, which is mainly caused by random training sampling (RTS) of selecting training data. This is because a different set of training samples may produce a different classification result. A general approach to addressing this problem is the so-called K-fold method which implements RTS K times and takes the average of overall accuracy with respect to standard deviation to describe a confidence level of classification performance. To deal with this issue, this article develops an iterative RTS (IRTS) method as an alternative to the K-fold method to reduce the uncertainty caused by RTS. Its idea is to add the spatial filtered classification maps to the image cube that is currently being processed via feedback loops to augment image cubes iteratively. Then, the training samples will be reselected randomly from the new augmented image cubes iteration-by-iteration. As a result, the training samples selected from each iteration will be updated by new added spatial information captured by spatial filters implemented at the iteration. The experimental results clearly demonstrate that IRTS successfully improves classification accuracy as well as reduces inconsistency in results. |
topic |
Hyperspectral image classification (HSIC) iterative random training sampling (IRTS) <named-content xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" content-type="math" xlink:type="simple"> <inline-formula> <tex-math notation="LaTeX">$K$</tex-math> </inline-formula> </named-content>-fold method random training sampling (RTS) spectral-spatial (SS) |
url |
https://ieeexplore.ieee.org/document/9137652/ |
work_keys_str_mv |
AT cheinichang iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages AT kennethyeonkongma iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages AT chiachenliang iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages AT yimeikuo iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages AT shuhanchen iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages AT shengweizhong iterativerandomtrainingsamplingspectralspatialclassificationforhyperspectralimages |
_version_ |
1721398832275652608 |