%0 Journal Article
%4 sid.inpe.br/mtc-m21c/2020/06.02.13.00
%2 sid.inpe.br/mtc-m21c/2020/06.02.13.00.32
%@doi 10.3390/RS12081284
%@issn 2072-4292
%T Comparison of cloud cover detection algorithms on sentinel-2 images of the Amazon tropical forest
%D 2020
%8 Apr.
%9 journal article
%A Sanchez Ipia, Alber Hamersson,
%A Picoli, Michelle Cristina Araújo,
%A Câmara, Gilberto,
%A Andrade Neto, Pedro Ribeiro de,
%A Chaves, Michel Eustáquio Dantas,
%A Lechler, Sarah,
%A Soares, Anderson Reis,
%A Marujo, Rennan de Freitas Bezerra,
%A Simões, Rolf Ezequiel de Oliveira,
%A Ferreira, Karine Reis,
%A Queiroz, Gilberto Ribeiro,
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation University of Münster
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@electronicmailaddress alber.ipia@inpe.br
%@electronicmailaddress michelle.picoli@inpe.br
%@electronicmailaddress gilberto.camara@inpe.br
%@electronicmailaddress pedro.andrade@inpe.br
%@electronicmailaddress michel.chaves@inpe.br
%@electronicmailaddress s.lechler@uni-muenster.de
%@electronicmailaddress anderson.soares@inpe.br
%@electronicmailaddress renan.marujo@inpe.br
%@electronicmailaddress rolf.simoes@inpe.br
%@electronicmailaddress karine.ferreira@inpe.br
%@electronicmailaddress gilberto.queiroz@inpe.br
%B Remote Sensing
%V 12
%N 8
%P e1284
%K remote sensing, amazon forest, clouds, Sentinel–2, Fmask, Sen2Cor, MAJA, s2cloudless.
%X Tropical forests regulate the global water and carbon cycles and also host most of the worlds biodiversity. Despite their importance, they are hard to survey due to their location, extent, and particularly, their cloud coverage. Clouds hinder the spatial and radiometric correction of satellite imagery and also diminishing the useful area on each image, making it difficult to monitor land change. For this reason, our purpose is to identify the cloud detection algorithm best suited for the Amazon rainforest on Sentinel2 images. To achieve this, we tested four cloud detection algorithms on Sentinel2 images spread in five areas of the Amazonia. Using more than eight thousand validation points, we compared four cloud detection methods: Fmask 4, MAJA, Sen2Cor, and s2cloudless. Our results point out that FMask 4 has the best overall accuracy on images of the Amazon region (90%), followed by Sen2Cors (79%), MAJA (69%), and S2cloudless (52%). We note the choice of method depends on the intended use. Since MAJA reduces the number of false positives by design, users that aim to improve the producers accuracy should consider its use.
%@language en
%3 sanchez_comparison.pdf