Identifying factors impacting the overall accuracy in image classification problems: a statistical approach
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Revista Tecnologia e Sociedade (Online) |
Texto Completo: | https://periodicos.utfpr.edu.br/rts/article/view/15480 |
Resumo: | Image classification is a subject of pattern recognition that can be applied in several areas. Obtaining highly-accurate classification involves choosing optimal set-ups from which images will be classified. In this process, controllable variables can affect the overall classification accuracy, such as the image’s spatial resolution and the classification method. In this sense, we have designed a factorial experiment where the classification accuracy of an image (from Curitiba, Paraná, Brazil) was obtained from three satellites and three classification methods. The Kruskal-Wallis test was applied to evaluate if the variability across factor levels supports the hypothesis that the experimental factors’ effects are statistically significant. Then, we evaluated which factor levels differed from each other using post-hoc tests. Our findings suggest that the image’s spatial resolution and the interaction between Satellite and Classification Method are determinants in obtaining accurate image classifications in a geographical context. |
id |
UTFPR-6_c5a6cbae9a0e97e5eda097b61b075611 |
---|---|
oai_identifier_str |
oai:periodicos.utfpr:article/15480 |
network_acronym_str |
UTFPR-6 |
network_name_str |
Revista Tecnologia e Sociedade (Online) |
repository_id_str |
|
spelling |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approachGeociências; Geofísica; Sensoriamento RemotoFactorial design; image classification; Kruskal-Wallis test; overall classification accuracy; spatial resolutionImage classification is a subject of pattern recognition that can be applied in several areas. Obtaining highly-accurate classification involves choosing optimal set-ups from which images will be classified. In this process, controllable variables can affect the overall classification accuracy, such as the image’s spatial resolution and the classification method. In this sense, we have designed a factorial experiment where the classification accuracy of an image (from Curitiba, Paraná, Brazil) was obtained from three satellites and three classification methods. The Kruskal-Wallis test was applied to evaluate if the variability across factor levels supports the hypothesis that the experimental factors’ effects are statistically significant. Then, we evaluated which factor levels differed from each other using post-hoc tests. Our findings suggest that the image’s spatial resolution and the interaction between Satellite and Classification Method are determinants in obtaining accurate image classifications in a geographical context.Universidade Tecnológica Federal do Paraná (UTFPR)CAPESBertoli, WesleyJunior, José MarcatoOliveira, Lucas Yuri Dutra2022-10-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://periodicos.utfpr.edu.br/rts/article/view/1548010.3895/rts.v18n54.15480Revista Tecnologia e Sociedade; v. 18, n. 54 (2022); 261-274Revista Tecnologia e Sociedade; v. 18, n. 54 (2022); 261-2741984-35261809-004410.3895/rts.v18n54reponame:Revista Tecnologia e Sociedade (Online)instname:Universidade Tecnológica Federal do Paraná (UTFPR)instacron:UTFPRenghttps://periodicos.utfpr.edu.br/rts/article/view/15480/9115Direitos autorais 2022 CC-BYhttp://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccess2024-05-01T15:48:26Zoai:periodicos.utfpr:article/15480Revistahttps://periodicos.ifrs.edu.br/index.php/tearPUBhttps://periodicos.utfpr.edu.br/rts/oai||rts-ct@utfpr.edu.br1984-35261809-0044opendoar:2024-05-01T15:48:26Revista Tecnologia e Sociedade (Online) - Universidade Tecnológica Federal do Paraná (UTFPR)false |
dc.title.none.fl_str_mv |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
title |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
spellingShingle |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach Bertoli, Wesley Geociências; Geofísica; Sensoriamento Remoto Factorial design; image classification; Kruskal-Wallis test; overall classification accuracy; spatial resolution |
title_short |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
title_full |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
title_fullStr |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
title_full_unstemmed |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
title_sort |
Identifying factors impacting the overall accuracy in image classification problems: a statistical approach |
author |
Bertoli, Wesley |
author_facet |
Bertoli, Wesley Junior, José Marcato Oliveira, Lucas Yuri Dutra |
author_role |
author |
author2 |
Junior, José Marcato Oliveira, Lucas Yuri Dutra |
author2_role |
author author |
dc.contributor.none.fl_str_mv |
CAPES |
dc.contributor.author.fl_str_mv |
Bertoli, Wesley Junior, José Marcato Oliveira, Lucas Yuri Dutra |
dc.subject.por.fl_str_mv |
Geociências; Geofísica; Sensoriamento Remoto Factorial design; image classification; Kruskal-Wallis test; overall classification accuracy; spatial resolution |
topic |
Geociências; Geofísica; Sensoriamento Remoto Factorial design; image classification; Kruskal-Wallis test; overall classification accuracy; spatial resolution |
description |
Image classification is a subject of pattern recognition that can be applied in several areas. Obtaining highly-accurate classification involves choosing optimal set-ups from which images will be classified. In this process, controllable variables can affect the overall classification accuracy, such as the image’s spatial resolution and the classification method. In this sense, we have designed a factorial experiment where the classification accuracy of an image (from Curitiba, Paraná, Brazil) was obtained from three satellites and three classification methods. The Kruskal-Wallis test was applied to evaluate if the variability across factor levels supports the hypothesis that the experimental factors’ effects are statistically significant. Then, we evaluated which factor levels differed from each other using post-hoc tests. Our findings suggest that the image’s spatial resolution and the interaction between Satellite and Classification Method are determinants in obtaining accurate image classifications in a geographical context. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-10-01 |
dc.type.none.fl_str_mv |
|
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://periodicos.utfpr.edu.br/rts/article/view/15480 10.3895/rts.v18n54.15480 |
url |
https://periodicos.utfpr.edu.br/rts/article/view/15480 |
identifier_str_mv |
10.3895/rts.v18n54.15480 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
https://periodicos.utfpr.edu.br/rts/article/view/15480/9115 |
dc.rights.driver.fl_str_mv |
Direitos autorais 2022 CC-BY http://creativecommons.org/licenses/by/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Direitos autorais 2022 CC-BY http://creativecommons.org/licenses/by/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Tecnológica Federal do Paraná (UTFPR) |
publisher.none.fl_str_mv |
Universidade Tecnológica Federal do Paraná (UTFPR) |
dc.source.none.fl_str_mv |
Revista Tecnologia e Sociedade; v. 18, n. 54 (2022); 261-274 Revista Tecnologia e Sociedade; v. 18, n. 54 (2022); 261-274 1984-3526 1809-0044 10.3895/rts.v18n54 reponame:Revista Tecnologia e Sociedade (Online) instname:Universidade Tecnológica Federal do Paraná (UTFPR) instacron:UTFPR |
instname_str |
Universidade Tecnológica Federal do Paraná (UTFPR) |
instacron_str |
UTFPR |
institution |
UTFPR |
reponame_str |
Revista Tecnologia e Sociedade (Online) |
collection |
Revista Tecnologia e Sociedade (Online) |
repository.name.fl_str_mv |
Revista Tecnologia e Sociedade (Online) - Universidade Tecnológica Federal do Paraná (UTFPR) |
repository.mail.fl_str_mv |
||rts-ct@utfpr.edu.br |
_version_ |
1800237663149621248 |