Land Use/Land Cover (LULC) classification based on SAR/Sentinel 1 image in Federal District, Brazil

• Autor(a) principal: Lima Ramos Barbosa, Felipe
• Data de Publicação: 2021
• Outros Autores: Fontes Guimarães, Renato, Abílio de Carvalho Júnior, Osmar, Arnaldo Trancoso Gomes, Roberto
• Tipo de documento: Artigo
• Idioma: por; eng
• Título da fonte: Sociedade & natureza (Online)
• Texto Completo: https://seer.ufu.br/index.php/sociedadenatureza/article/view/55954
• topic: SAR; Sentinel 1; Coerência interferométrica; Ambiente urbano; Objetos urbanos; SAR; Sentinel 1; Interferometric coherence; Urban environment; Urban objects

The social and environmental challenges are directly related to theexisting population concentration in urban environments, whichcontribute to more than 75% of the world’s Gross Domestic Product(GDP). Thus, it is essential to detect the dynamics of Land Use andLand Cover (LULC) aiming to support public policies elaboration andimplementation. SAR systems, especially the interferometrytechniques, have been shown great results in face of this challenge,since they do not have direct influence from the atmosphere. Limitedstudies were conducted using interferometric coherence from the Sentinel-1 satellite in an urban environment. In this sense, theobjective of this study was to classify the LULC of part of the FederalDistrict, Brazil based on different dimensions considering the measuresof intensity and interferometric coherence for the year 2018. The resultsmeasured from the Kappa and F1 metrics indicate that the insertion ofa time series of interferometric coherencies improves the performance ofthe classification, from 0.50 to 0.75 (Kappa) and from 0.54to 0.79 (F1), afact that was evident in the improved performance of the thematicclasses related to vegetation cover. Furthermore, it is also found thatthe identification of urban objects is best represented by the use of onlythe intensities (VV and VH) in the classification process.