Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods
The data classification of fully polarimetric synthetic aperture radar (PolSAR) is one of the favourite topics in the remote sensing community. To date, a wide variety of algorithms have been utilized for PolSAR data classification, and among them kernel methods are the most attractive algorithms fo...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Taylor & Francis Group
2021-01-01
|
Series: | European Journal of Remote Sensing |
Subjects: | |
Online Access: | http://dx.doi.org/10.1080/22797254.2021.1924081 |
id |
doaj-16b7549fda9f47b68a93115fd58b1919 |
---|---|
record_format |
Article |
spelling |
doaj-16b7549fda9f47b68a93115fd58b19192021-08-09T18:41:15ZengTaylor & Francis GroupEuropean Journal of Remote Sensing2279-72542021-01-0154131031710.1080/22797254.2021.19240811924081Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methodsIman Khosravi0Yury Razoumny1Javad Hatami Afkoueieh2Seyed Kazem Alavipanah3University of TehranAcademy of Engineering, Peoples’ Friendship University of Russia (RUDN University)Academy of Engineering, Peoples’ Friendship University of Russia (RUDN University)University of TehranThe data classification of fully polarimetric synthetic aperture radar (PolSAR) is one of the favourite topics in the remote sensing community. To date, a wide variety of algorithms have been utilized for PolSAR data classification, and among them kernel methods are the most attractive algorithms for this purpose. The most famous kernel method, i.e., the support vector machine (SVM) has been widely used for PolSAR data classification. However, until now, no studies to classify PolSAR data have been carried out using certain extended SVM versions, such as the least squares support vector machine (LSSVM), relevance vector machine (RVM) and import vector machine (IVM). Therefore, this work has employed and compared these four kernel methods for the classification of three PolSAR data sets. These methods were compared in two groups: the SVM and LSSVM as non-probabilistic kernel methods vs. the RVM and IVM as probabilistic kernel methods. In general, the results demonstrated that the SVM was marginally better, more accurate and more stable than the probabilistic kernels. Furthermore, the LSSVM performed much faster than the probabilistic kernel methods and its associated version, the SVM, with comparable accuracy.http://dx.doi.org/10.1080/22797254.2021.1924081fully polarimetric sarclassificationkernel methodssvm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Iman Khosravi Yury Razoumny Javad Hatami Afkoueieh Seyed Kazem Alavipanah |
spellingShingle |
Iman Khosravi Yury Razoumny Javad Hatami Afkoueieh Seyed Kazem Alavipanah Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods European Journal of Remote Sensing fully polarimetric sar classification kernel methods svm |
author_facet |
Iman Khosravi Yury Razoumny Javad Hatami Afkoueieh Seyed Kazem Alavipanah |
author_sort |
Iman Khosravi |
title |
Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
title_short |
Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
title_full |
Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
title_fullStr |
Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
title_full_unstemmed |
Fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
title_sort |
fully polarimetric synthetic aperture radar data classification using probabilistic and non-probabilistic kernel methods |
publisher |
Taylor & Francis Group |
series |
European Journal of Remote Sensing |
issn |
2279-7254 |
publishDate |
2021-01-01 |
description |
The data classification of fully polarimetric synthetic aperture radar (PolSAR) is one of the favourite topics in the remote sensing community. To date, a wide variety of algorithms have been utilized for PolSAR data classification, and among them kernel methods are the most attractive algorithms for this purpose. The most famous kernel method, i.e., the support vector machine (SVM) has been widely used for PolSAR data classification. However, until now, no studies to classify PolSAR data have been carried out using certain extended SVM versions, such as the least squares support vector machine (LSSVM), relevance vector machine (RVM) and import vector machine (IVM). Therefore, this work has employed and compared these four kernel methods for the classification of three PolSAR data sets. These methods were compared in two groups: the SVM and LSSVM as non-probabilistic kernel methods vs. the RVM and IVM as probabilistic kernel methods. In general, the results demonstrated that the SVM was marginally better, more accurate and more stable than the probabilistic kernels. Furthermore, the LSSVM performed much faster than the probabilistic kernel methods and its associated version, the SVM, with comparable accuracy. |
topic |
fully polarimetric sar classification kernel methods svm |
url |
http://dx.doi.org/10.1080/22797254.2021.1924081 |
work_keys_str_mv |
AT imankhosravi fullypolarimetricsyntheticapertureradardataclassificationusingprobabilisticandnonprobabilistickernelmethods AT yuryrazoumny fullypolarimetricsyntheticapertureradardataclassificationusingprobabilisticandnonprobabilistickernelmethods AT javadhatamiafkoueieh fullypolarimetricsyntheticapertureradardataclassificationusingprobabilisticandnonprobabilistickernelmethods AT seyedkazemalavipanah fullypolarimetricsyntheticapertureradardataclassificationusingprobabilisticandnonprobabilistickernelmethods |
_version_ |
1721213610639753216 |