Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica
For decades the detection and monitoring of land cover have been performed by aerial-transported remote sensing and satellites. Detecting and quantifying land cover relays on sensor capacities and classification techniques, such as supervised, unsupervised and mixed. The supervised method is the mos...
Autores Principales: | Vargas-Sanabria, Daniela, Campos-Vargas, Carlos |
---|---|
Formato: | Artículo |
Idioma: | Español |
Publicado: |
Editorial Tecnológica de Costa Rica (entidad editora)
2018
|
Materias: | |
Acceso en línea: |
https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/3497 https://hdl.handle.net/2238/11781 |
id |
RepoTEC11781 |
---|---|
recordtype |
dspace |
spelling |
RepoTEC117812020-09-25T23:11:59Z Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica Sistema multi-algoritmo para la clasificación de coberturas de la tierra en el bosque seco tropical del Área de Conservación Guanacaste, Costa Rica Vargas-Sanabria, Daniela Campos-Vargas, Carlos Multi-algorithm land cover remote sensing supervised classification. Multi-algoritmo cobertura de la tierra teledetección clasificación supervisada. For decades the detection and monitoring of land cover have been performed by aerial-transported remote sensing and satellites. Detecting and quantifying land cover relays on sensor capacities and classification techniques, such as supervised, unsupervised and mixed. The supervised method is the most accurate and depends on the ability of the algorithm used to discriminate the categories. On the land cover associated to the tropical dry forest at Guanacaste Conservation Area, Costa Rica, supervised classifications were used, using the Minimum Distance, Mahalanobies, Maximum Likelihood, Neural Network, Support Vector Machine and Parallelepiped algorithms to determine which algorithm classified the coverages better (late forest, early forest, intermediate forest, gallery forest, pasture, mangrove) according to control points taken through field work. According to the kappa index and precision values the performance of Maximum Likelihood and Neural Network was highlighted in the classification of land coverages. This study demonstrates that a hierarchical and multi-algorithm scheme can propel the results of the classification of land cover by considering the advantages and limitations of each classification algorithm, especially when considering aspects related to sample size, sensor resolutions (temporal, spatial, radiometric, spectral), atmospheric conditions and composition of vegetation and landscape. Por décadas la detección y el monitoreo de coberturas de la tierra se ha llevado a cabo por medio de teledetección aéreo-transportada y satelital. La habilidad de detectar y cuantificar coberturas de la tierra depende en gran medida de las capacidades del sensor y técnicas de clasificación entre las cuales destacan supervisada, no supervisada y mixta. El método supervisado se considera el más preciso; sin embargo, depende de la capacidad del algoritmo utilizado para discriminar las categorías. En las coberturas de la tierra asociadas al bosque seco tropical del Área de Conservación Guanacaste, Costa Rica se realizaron series de clasificaciones supervisadas, utilizando los algoritmos Minimum Distance, Mahalanobies, Maximum Likelihood, Neural Network, Support Vector Machine y Parallelepiped para determinar cuál algoritmo clasificaba mejor cada cobertura (bosque tardío, bosque temprano, bosque intermedio, bosque de galería, pastos, manglar) según puntos de control tomados mediante trabajo de campo. De acuerdo al índice de kappa y valores de precisión se destacó el rendimiento de Maximum Likelihood y Neural Network en la clasificación de las coberturas de la tierra. Este estudio demuestra que un esquema jerárquico y multi-algoritmo puede propulsar los resultados de la clasificación de coberturas de la tierra al contemplar las ventajas y limitaciones de cada algoritmo de clasificación; especialmente si se considera aspectos relativos al tamaño de muestra, resoluciones del sensor (temporal, espacial, radiométrica, espectral), condiciones atmosféricas y composición de la vegetación y paisaje. 2018-03-22 2020-09-25T23:11:59Z 2020-09-25T23:11:59Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/3497 10.18845/tm.v31i1.3497 https://hdl.handle.net/2238/11781 spa https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/3497/pdf application/pdf Editorial Tecnológica de Costa Rica (entidad editora) Tecnología en marcha Journal; Vol. 31 No. 1: Enero-Marzo 2018; 58-69 Revista Tecnología en Marcha; Vol. 31 Núm. 1: Enero-Marzo 2018; 58-69 2215-3241 0379-3982 |
institution |
Tecnológico de Costa Rica |
collection |
Repositorio TEC |
language |
Español |
topic |
Multi-algorithm land cover remote sensing supervised classification. Multi-algoritmo cobertura de la tierra teledetección clasificación supervisada. |
spellingShingle |
Multi-algorithm land cover remote sensing supervised classification. Multi-algoritmo cobertura de la tierra teledetección clasificación supervisada. Vargas-Sanabria, Daniela Campos-Vargas, Carlos Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
description |
For decades the detection and monitoring of land cover have been performed by aerial-transported remote sensing and satellites. Detecting and quantifying land cover relays on sensor capacities and classification techniques, such as supervised, unsupervised and mixed. The supervised method is the most accurate and depends on the ability of the algorithm used to discriminate the categories. On the land cover associated to the tropical dry forest at Guanacaste Conservation Area, Costa Rica, supervised classifications were used, using the Minimum Distance, Mahalanobies, Maximum Likelihood, Neural Network, Support Vector Machine and Parallelepiped algorithms to determine which algorithm classified the coverages better (late forest, early forest, intermediate forest, gallery forest, pasture, mangrove) according to control points taken through field work. According to the kappa index and precision values the performance of Maximum Likelihood and Neural Network was highlighted in the classification of land coverages. This study demonstrates that a hierarchical and multi-algorithm scheme can propel the results of the classification of land cover by considering the advantages and limitations of each classification algorithm, especially when considering aspects related to sample size, sensor resolutions (temporal, spatial, radiometric, spectral), atmospheric conditions and composition of vegetation and landscape. |
format |
Artículo |
author |
Vargas-Sanabria, Daniela Campos-Vargas, Carlos |
author_sort |
Vargas-Sanabria, Daniela |
title |
Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
title_short |
Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
title_full |
Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
title_fullStr |
Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
title_full_unstemmed |
Multi-algorithm system for land cover classification in tropical dry forest at Guanacaste Conservation Area, Costa Rica |
title_sort |
multi-algorithm system for land cover classification in tropical dry forest at guanacaste conservation area, costa rica |
publisher |
Editorial Tecnológica de Costa Rica (entidad editora) |
publishDate |
2018 |
url |
https://revistas.tec.ac.cr/index.php/tec_marcha/article/view/3497 https://hdl.handle.net/2238/11781 |
_version_ |
1796138820391927808 |
score |
12.238652 |