Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes

C.H.W Souza, E. Mercante, V.H.R. Prudente and D.D.D. Justina. 2013. Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes. Cien. Inv. Agr. 40(2): 419-428. Satellite imagery, in combination with remote sensing techniques, provides a...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Souza,Carlos H. Wachholz de, Mercante,Erivelto, Prudente,Victor H. R., Justina,Diego D.D.
Lenguaje:	English
Publicado:	Pontificia Universidad Católica de Chile. Facultad de Agronomía e Ingeniería Forestal 2013
Materias:	Accuracy indices agricultural landscape classifiers remote sensing
Acceso en línea:	http://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-16202013000200016
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:scielo:S0718-16202013000200016
record_format	dspace
spelling	oai:scielo:S0718-162020130002000162013-10-23Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classesSouza,Carlos H. Wachholz deMercante,EriveltoPrudente,Victor H. R.Justina,Diego D.D. Accuracy indices agricultural landscape classifiers remote sensing C.H.W Souza, E. Mercante, V.H.R. Prudente and D.D.D. Justina. 2013. Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes. Cien. Inv. Agr. 40(2): 419-428. Satellite imagery, in combination with remote sensing techniques, provides a new opportunity for monitoring and assessing crops with lower cost and greater objectivity than traditional surveys. The present research employed Landsat 5/TM satellite imagery to identify the land cover classes in Cafelândia (Paraná, Brasil), a predominantly agricultural town. Five supervised classification methods (parallelepiped (PL), minimum distance (MND), Mahalanobis distance (MHD), maximum likelihood classifier (MLC) and spectral angle mapper (SAM)) were tested in this work. To assess the efficiency of the classifications, accuracy indices and error metrics obtained through total confusion matrices were used. The results indicated that the Mahalanobis and SAM methods generated the smallest errors for the four studied land use classes (soybean, corn, forest, and bare soil), with overall accuracy values of 88% and 86%, respectively, and kappa index values 0.83 and 0.80, respectively. The values of these methods for the applied metrics were 0.88 and 0.86 for the sensitivity index, 0.96 and 0.95 for the total specificity index and 0.84 and 0.81 for Matthews correlation coefficient, respectively. The different classification methods clearly exhibited large variations in their performance for land cover mapping. The use of measures obtained from the error matrix is a suitable method for comparisons of thematic maps.info:eu-repo/semantics/openAccessPontificia Universidad Católica de Chile. Facultad de Agronomía e Ingeniería ForestalCiencia e investigación agraria v.40 n.2 20132013-05-01text/htmlhttp://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-16202013000200016en10.4067/S0718-16202013000200016
institution	Scielo Chile
collection	Scielo Chile
language	English
topic	Accuracy indices agricultural landscape classifiers remote sensing
spellingShingle	Accuracy indices agricultural landscape classifiers remote sensing Souza,Carlos H. Wachholz de Mercante,Erivelto Prudente,Victor H. R. Justina,Diego D.D. Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
description	C.H.W Souza, E. Mercante, V.H.R. Prudente and D.D.D. Justina. 2013. Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes. Cien. Inv. Agr. 40(2): 419-428. Satellite imagery, in combination with remote sensing techniques, provides a new opportunity for monitoring and assessing crops with lower cost and greater objectivity than traditional surveys. The present research employed Landsat 5/TM satellite imagery to identify the land cover classes in Cafelândia (Paraná, Brasil), a predominantly agricultural town. Five supervised classification methods (parallelepiped (PL), minimum distance (MND), Mahalanobis distance (MHD), maximum likelihood classifier (MLC) and spectral angle mapper (SAM)) were tested in this work. To assess the efficiency of the classifications, accuracy indices and error metrics obtained through total confusion matrices were used. The results indicated that the Mahalanobis and SAM methods generated the smallest errors for the four studied land use classes (soybean, corn, forest, and bare soil), with overall accuracy values of 88% and 86%, respectively, and kappa index values 0.83 and 0.80, respectively. The values of these methods for the applied metrics were 0.88 and 0.86 for the sensitivity index, 0.96 and 0.95 for the total specificity index and 0.84 and 0.81 for Matthews correlation coefficient, respectively. The different classification methods clearly exhibited large variations in their performance for land cover mapping. The use of measures obtained from the error matrix is a suitable method for comparisons of thematic maps.
author	Souza,Carlos H. Wachholz de Mercante,Erivelto Prudente,Victor H. R. Justina,Diego D.D.
author_facet	Souza,Carlos H. Wachholz de Mercante,Erivelto Prudente,Victor H. R. Justina,Diego D.D.
author_sort	Souza,Carlos H. Wachholz de
title	Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
title_short	Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
title_full	Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
title_fullStr	Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
title_full_unstemmed	Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
title_sort	methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes
publisher	Pontificia Universidad Católica de Chile. Facultad de Agronomía e Ingeniería Forestal
publishDate	2013
url	http://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-16202013000200016
work_keys_str_mv	AT souzacarloshwachholzde methodsofperformanceevaluationforthesupervisedclassificationofsatelliteimageryindetermininglandcoverclasses AT mercanteerivelto methodsofperformanceevaluationforthesupervisedclassificationofsatelliteimageryindetermininglandcoverclasses AT prudentevictorhr methodsofperformanceevaluationforthesupervisedclassificationofsatelliteimageryindetermininglandcoverclasses AT justinadiegodd methodsofperformanceevaluationforthesupervisedclassificationofsatelliteimageryindetermininglandcoverclasses
_version_	1714202150776602624

Methods of performance evaluation for the supervised classification of satellite imagery in determining land cover classes

Ejemplares similares