@InProceedings{PrudenteViMoOlSaAd:2019:UtDaSA,
author = "Prudente, Victor Hugo Rohden and Vieira, Denis Corte and
Montibeller, Bruno and Oldoni, Lucas Volochen and Sanches, Ieda
Del'Arco and Adami, Marcos",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {University of Tartu}
and {Instituto Nacional de Pesquisas Espaciais (INPE)} and
{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)}",
title = "Utiliza{\c{c}}{\~a}o de dados SAR na classifica{\c{c}}{\~a}o
de esp{\'e}cies agr{\'{\i}}colas de primeira e segunda safra",
booktitle = "Anais...",
year = "2019",
editor = "Gherardi, Douglas Francisco Marcolino and Sanches, Ieda DelArco
and Arag{\~a}o, Luiz Eduardo Oliveira e Cruz de",
pages = "1643--1646",
organization = "Simp{\'o}sio Brasileiro de Sensoriamento Remoto, 19. (SBSR)",
publisher = "Instituto Nacional de Pesquisas Espaciais (INPE)",
address = "S{\~a}o Jos{\'e} dos Campos",
keywords = "Random Forest, microondas, banda C, agricultura, Random Forest,
microwave, C-band, agriculture.",
abstract = "Dados dos sensores SAR possuem a vantagem de serem menos
influenciados por presen{\c{c}}a de nuvens, possibilitando maior
frequ{\^e}ncia temporal para monitoramento de alvos
agr{\'{\i}}colas. Diante disso, o objetivo deste trabalho foi
utilizar dados SAR para discriminar diferentes esp{\'e}cies
agr{\'{\i}}colas e alvos naturais no munic{\'{\i}}pio de Luiz
Eduardo Magalh{\~a}es Bahia, em duas safras. Foi utilizada
abordagem temporal, com 12 imagens SAR/Sentinel-1tanto para a
segunda safra de 2017 quanto para a primeira safra de 2018. O
classificar utilizado foi o Random Forest e as amostras de
treinamento foram adquiridas em visitas de campo. Entre os
resultados, a classifica{\c{c}}{\~a}o do algod{\~a}o obteve as
melhores acur{\'a}cias para ambas as safras. Na segunda safra de
2017 houve confus{\~a}o entre as classes de milheto, milho e
sorgo e entre as classes de eucalipto, caf{\'e} e cerrado,
al{\'e}m das classes grama e pastagem. Para a primeira safra de
2018 obteve-se acur{\'a}cias melhores para separa{\c{c}}{\~a}o
das esp{\'e}cies agr{\'{\i}}colas. ABSTRACT: SAR sensors data
have the advantage of being less influenced by the presence of
clouds, thus allowing higher temporal frequency for the monitoring
of agricultural targets. Therefore, the objective of this work was
to use SAR data to classify different agricultural species and
natural targets in the municipality of Luiz Eduardo Magalh{\~a}es
Bahia, in two harvests. Temporal approach was used, with 12 SAR /
Sentinel-1 images for the second harvest of 2017 and for the first
harvest of 2018. The Random Forest algorithm was used, and the
training samples were acquired at field visits. The classification
of Cotton obtained the best results for both harvests. In the
second crop of 2017 there was confusion among the classes of
millet, corn and sorghum and among the classes of eucalyptus,
coffee and cerrado, besides the grass and pasture classes. Better
accuracy for the separation of agricultural species was obtained
for the first harvest of 2018.",
conference-location = "Santos",
conference-year = "14-17 abril 2019",
isbn = "978-85-17-00097-3",
language = "pt",
organisation = "Instituto Nacional de Pesquisas Espaciais (INPE)",
ibi = "8JMKD3MGP6W34M/3U9T4L5",
url = "http://urlib.net/ibi/8JMKD3MGP6W34M/3U9T4L5",
targetfile = "97850.pdf",
type = "Produ{\c{c}}{\~a}o e previs{\~a}o agr{\'{\i}}cola",
urlaccessdate = "27 abr. 2024"
}