Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances
Autor(a) principal: | |
---|---|
Data de Publicação: | 2018 |
Outros Autores: | , , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1080/17538947.2018.1474958 http://hdl.handle.net/11449/171073 |
Resumo: | Region-based classification of PolSAR data can be effectively performed by seeking for the assignment that minimizes a distance between prototypes and segments. Silva et al. [“Classification of segments in PolSAR imagery by minimum stochastic distances between wishart distributions.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 6 (3): 1263–1273] used stochastic distances between complex multivariate Wishart models which, differently from other measures, are computationally tractable. In this work we assess the robustness of such approach with respect to errors in the training stage, and propose an extension that alleviates such problems. We introduce robustness in the process by incorporating a combination of radial basis kernel functions and stochastic distances with Support Vector Machines (SVM). We consider several stochastic distances between Wishart: Bhatacharyya, Kullback-Leibler, Chi-Square, Rényi, and Hellinger. We perform two case studies with PolSAR images, both simulated and from actual sensors, and different classification scenarios to compare the performance of Minimum Distance and SVM classification frameworks. With this, we model the situation of imperfect training samples. We show that SVM with the proposed kernel functions achieves better performance with respect to Minimum Distance, at the expense of more computational resources and the need of parameter tuning. Code and data are provided for reproducibility. |
id |
UNSP_5f5f37fd64b66de6963101f9d35198e9 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/171073 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distancesimage classificationminimum distance classifierPolSARstochastic distanceSVMRegion-based classification of PolSAR data can be effectively performed by seeking for the assignment that minimizes a distance between prototypes and segments. Silva et al. [“Classification of segments in PolSAR imagery by minimum stochastic distances between wishart distributions.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 6 (3): 1263–1273] used stochastic distances between complex multivariate Wishart models which, differently from other measures, are computationally tractable. In this work we assess the robustness of such approach with respect to errors in the training stage, and propose an extension that alleviates such problems. We introduce robustness in the process by incorporating a combination of radial basis kernel functions and stochastic distances with Support Vector Machines (SVM). We consider several stochastic distances between Wishart: Bhatacharyya, Kullback-Leibler, Chi-Square, Rényi, and Hellinger. We perform two case studies with PolSAR images, both simulated and from actual sensors, and different classification scenarios to compare the performance of Minimum Distance and SVM classification frameworks. With this, we model the situation of imperfect training samples. We show that SVM with the proposed kernel functions achieves better performance with respect to Minimum Distance, at the expense of more computational resources and the need of parameter tuning. Code and data are provided for reproducibility.UNESP – Universidade Estadual Paulista, ICT – Instituto de Ciência e Tecnologia, São José dos Campos, BrazilLaCCAN – Laboratório de de Computação Científica e Análise Numérica, UFAL – Universidade Federal de Alagoas, Maceió, BrazilSeção de Ensino de Engenharia Cartográfica, IME – Instituto Militar de Engenharia, Rio de Janeiro, BrazilDPI – Divisão de Processamento de Imagens, INPE – Instituto Nacional de Pesquisas Espaciais, São José dos Campos, BrazilUniversidade Estadual Paulista (Unesp)Negri, Rogério G.Frery, Alejandro C.Silva, Wagner B.Mendes, Tatiana S. G.Dutra, Luciano V.2018-12-11T16:53:38Z2018-12-11T16:53:38Z2018-06-05info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/article1-21application/pdfhttp://dx.doi.org/10.1080/17538947.2018.1474958International Journal of Digital Earth, p. 1-21.1753-89551753-8947http://hdl.handle.net/11449/17107310.1080/17538947.2018.14749582-s2.0-850479388942-s2.0-85047938894.pdf82018051329812880000-0002-4808-2362Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengInternational Journal of Digital Earth0,7280,728info:eu-repo/semantics/openAccess2023-10-19T06:03:53Zoai:repositorio.unesp.br:11449/171073Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462023-10-19T06:03:53Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
title |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
spellingShingle |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances Negri, Rogério G. image classification minimum distance classifier PolSAR stochastic distance SVM |
title_short |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
title_full |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
title_fullStr |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
title_full_unstemmed |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
title_sort |
Region-based classification of PolSAR data using radial basis kernel functions with stochastic distances |
author |
Negri, Rogério G. |
author_facet |
Negri, Rogério G. Frery, Alejandro C. Silva, Wagner B. Mendes, Tatiana S. G. Dutra, Luciano V. |
author_role |
author |
author2 |
Frery, Alejandro C. Silva, Wagner B. Mendes, Tatiana S. G. Dutra, Luciano V. |
author2_role |
author author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) |
dc.contributor.author.fl_str_mv |
Negri, Rogério G. Frery, Alejandro C. Silva, Wagner B. Mendes, Tatiana S. G. Dutra, Luciano V. |
dc.subject.por.fl_str_mv |
image classification minimum distance classifier PolSAR stochastic distance SVM |
topic |
image classification minimum distance classifier PolSAR stochastic distance SVM |
description |
Region-based classification of PolSAR data can be effectively performed by seeking for the assignment that minimizes a distance between prototypes and segments. Silva et al. [“Classification of segments in PolSAR imagery by minimum stochastic distances between wishart distributions.” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 6 (3): 1263–1273] used stochastic distances between complex multivariate Wishart models which, differently from other measures, are computationally tractable. In this work we assess the robustness of such approach with respect to errors in the training stage, and propose an extension that alleviates such problems. We introduce robustness in the process by incorporating a combination of radial basis kernel functions and stochastic distances with Support Vector Machines (SVM). We consider several stochastic distances between Wishart: Bhatacharyya, Kullback-Leibler, Chi-Square, Rényi, and Hellinger. We perform two case studies with PolSAR images, both simulated and from actual sensors, and different classification scenarios to compare the performance of Minimum Distance and SVM classification frameworks. With this, we model the situation of imperfect training samples. We show that SVM with the proposed kernel functions achieves better performance with respect to Minimum Distance, at the expense of more computational resources and the need of parameter tuning. Code and data are provided for reproducibility. |
publishDate |
2018 |
dc.date.none.fl_str_mv |
2018-12-11T16:53:38Z 2018-12-11T16:53:38Z 2018-06-05 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1080/17538947.2018.1474958 International Journal of Digital Earth, p. 1-21. 1753-8955 1753-8947 http://hdl.handle.net/11449/171073 10.1080/17538947.2018.1474958 2-s2.0-85047938894 2-s2.0-85047938894.pdf 8201805132981288 0000-0002-4808-2362 |
url |
http://dx.doi.org/10.1080/17538947.2018.1474958 http://hdl.handle.net/11449/171073 |
identifier_str_mv |
International Journal of Digital Earth, p. 1-21. 1753-8955 1753-8947 10.1080/17538947.2018.1474958 2-s2.0-85047938894 2-s2.0-85047938894.pdf 8201805132981288 0000-0002-4808-2362 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
International Journal of Digital Earth 0,728 0,728 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
1-21 application/pdf |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1799964614913425408 |