@InProceedings{DallCortivoKamp:2023:EsDeRe,
               author = "Dall Cortivo, F{\'a}bio and Kampel, Milton",
          affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto 
                         Nacional de Pesquisas Espaciais (INPE)}",
                title = "Estudo do desempenho de redes neurais artificiais na 
                         constru{\c{c}}{\~a}o de s{\'e}ries longas de reflect{\^a}ncia 
                         de sensoriamento remoto",
            booktitle = "Anais...",
                 year = "2023",
               editor = "Gherardi, Douglas Francisco Marcolino and Arag{\~a}o, Luiz 
                         Eduardo Oliveira e Cruz de and Sanches, Ieda DelArco",
                pages = "e156068",
         organization = "Simp{\'o}sio Brasileiro de Sensoriamento Remoto, 20. (SBSR)",
            publisher = "Instituto Nacional de Pesquisas Espaciais (INPE)",
              address = "S{\~a}o Jos{\'e} dos Campos",
             keywords = "intelig{\^e}ncia artificial, redes neurais, s{\'e}ries longas, 
                         reflect{\^a}ncia de sensoriamento remoto, cor do oceano, 
                         artificial inteligence, neural networks, long data series, remote 
                         sensing reflectance, ocean color.",
             abstract = "Esse trabalho visa avaliar o desempenho de redes neurais 
                         artificiais quando aplicadas com a finalidade de se obter 
                         s{\'e}ries temporais longas para reflect{\^a}ncia de 
                         sensoriamento remoto (RRS). Foi utilizado o per{\'{\i}}odo de 
                         sobre posi{\c{c}}{\~a}o das imagens (2002-2006) SeaWiFS e 
                         Modis/Aqua, na regi{\~a}o de plataforma da Bacia de Santos, para 
                         treinar uma rede neural do tipo Perceptron de M{\'u}ltiplas 
                         Camadas, a fim de converter as RRS nas bandas vis{\'{\i}}vel do 
                         sensor SeaWiFS em RRS Modis/Aqua nas bandas 443, 488 e 547, que 
                         s{\~a}o as bandas Modis/Aqua utilizadas no algoritmo OC3M para 
                         estimativa de clorofila. Os resultados apresentados avaliam o 
                         desempenho da rede na convers{\~a}o das reflect{\^a}ncias 
                         SeaWiFS em reflect{\^a}ncias Modis/Aqua para o per{\'{\i}}odo 
                         de sobreposi{\c{c}}{\~a}o. Para esta valida{\c{c}}{\~a}o foi 
                         comparado a RRS Modis/Aqua com a RRS dada pela rede. Os resultados 
                         mostram R2 0; 80 e correla{\c{c}}{\~a}o entre as s{\'e}ries 
                         (para cada banda) superior a 0,9. ABSTRACT: This work aims to 
                         evaluate the performance of artificial neural networks when 
                         applied in order to obtain long time series for remote sensing 
                         reflectance (RRS). The overlapping period of the images 
                         (2002-2006) SeaWiFS and Modis/Aqua, in the shelf region of the 
                         Santos Basin, was used to train a neural network of the 
                         Multi-Layer Perceptron type, in order to convert the RRS in 
                         visible bands of the SeaWiFS sensor into RRS Modia/Aqua in bands 
                         443, 488 and 547. These Modis/Aqua bands are used in the OC3M 
                         algorithm for chlorophyll estimation. The presented results 
                         evaluate the performance of the network in the conversion of 
                         SeaWiFS reflectances into Modis/Aqua reflectances for the 
                         overlapping period. For this validation, the RRS Modis/Aqua was 
                         compared with the RRS given by the network. The results show R2 0; 
                         80 and a correlation between the series (for each band) greater 
                         than 0; 9.",
  conference-location = "Florian{\'o}polis",
      conference-year = "02-05 abril 2023",
                 isbn = "978-65-89159-04-9",
             language = "pt",
         organisation = "Instituto Nacional de Pesquisas Espaciais (INPE)",
                  ibi = "8JMKD3MGP6W34M/48UM9AL",
                  url = "http://urlib.net/ibi/8JMKD3MGP6W34M/48UM9AL",
           targetfile = "156068.pdf",
                 type = "Intelig{\^e}ncia Artificial para Observa{\c{c}}{\~a}o da 
                         Terra",
        urlaccessdate = "23 maio 2024"
}