Análisis de uso de suelo a partir de imágenes satelitales sentinel 2 en el cantón Buena Fe, provincia de Los Ríos

This research shows an analysis of land use from Sentinel 2 satellite images in the Buena Fe canton, province of Los Ríos, whose multispectral resolution is 10 meters, which allowed determining with greater accuracy and detail the calculation of spectral indices such as NDVI, SAVI and NDMI that show the characteristics of the vegetation cover. The maximum likelihood method (Maximum Likelihood) allowed us to perform the supervised classification with which we identified the types of land cover and land use in the study area. Six categories were identified, whose percentages of coverage marked the bush zone with 37% coverage, followed by cocoa and banana crops with 29% and 20% respectively, while the lowest percentages were for palm crops with 9% and water bodies with 4%. Finally, to check the accuracy of the methods applied and validate the classification, a confusion matrix was calculated, resulting in a double-entry table with an accuracy of 95.9% efficiency. The calculation of the kappa coefficient was also applied to evaluate the level of satisfaction of the results, obtaining a score of 0.94, which according to the table proposed by Cohen means that the classification is very good with a high degree of acceptance. Keywords: Land use, land cover, Sentinel 2 satellite images, supervised classification, spectral indices, spatial distribution.

• Autor principal: Vera Rojas, Nolberto Javier (author)
• Format: bachelorThesis
• Idioma: spa
• Publicat: 2020
• Matèries: Uso de suelo; Cobertura de suelo; Imágenes satelitales Sentinel 2; Clasificación supervisada
• Accés en línia: http://repositorio.uteq.edu.ec/handle/43000/6186