@InProceedings{MatosSáncCarn:2022:DeLeAu,
author = "Matos, Diego Henrique M. and S{\'a}nchez Ipia, Alber Hamersson
and Carneiro, Tiago G. S.",
affiliation = "{Universidade Federal de Minas Gerais (UFMG)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {Universidade Federal
de Ouro Preto (UFOP)}",
title = "Deep learning automated workflow for cloud segmentation in remote
sensing images",
booktitle = "Anais...",
year = "2022",
editor = "Rosim, Sergio (INPE) and Santos, Leonardo Bacelar Lima (CEMADEN)
and Pereira, Marconi de Arruda (UFSJ)",
organization = "Simp{\'o}sio Brasileiro de Geoinform{\'a}tica, 23. (GEOINFO)",
publisher = "Instituto Nacional de Pesquisas Espaciais (INPE)",
address = "S{\~a}o Jos{\'e} dos Campos",
abstract = "In this work, we propose an open source and automated workflow for
semantic segmentation of remote sensing images. Even though it can
be used in other sensors, to evaluate this workflow, a case study
has been conducted applying a deep learning algorithm for
segmenting clouds in images from WFI sensor, onboard CBERS-4A
satellite. Since WFI does not have a tailor-made cloud
segmentation algorithm, we customized our workflow based on the
Unet neural network to fulfill this gap. Our results are promising
according to our tests, although some problems were identified,
like false positives over high albedo targets. These problems
suggest improvements that could be tackled in the future.",
conference-location = "On-line",
conference-year = "28 a 30 nov. 2022",
issn = "2179-4847",
language = "en",
ibi = "8JMKD3MGPDW34P/487M6K5",
url = "http://urlib.net/ibi/8JMKD3MGPDW34P/487M6K5",
targetfile = "192-203_Matos_deep.pdf",
urlaccessdate = "17 maio 2024"
}