%0 Conference Proceedings
%4 sid.inpe.br/marte2/2017/10.27.13.47.54
%2 sid.inpe.br/marte2/2017/10.27.13.47.55
%@isbn 978-85-17-00088-1
%F 59884
%T Interferência do método de aquisição e processamento de fotos hemisféricas na determinação do Índice de Área Foliar: um estudo preliminar
%D 2017
%A Silva, Daniel Augusto da,
%A Vibrans, Alexander Christian,
%@electronicmailaddress daniel.florestal@live.com
%E Gherardi, Douglas Francisco Marcolino,
%E Aragão, Luiz Eduardo Oliveira e Cruz de,
%B Simpósio Brasileiro de Sensoriamento Remoto, 18 (SBSR)
%C Santos
%8 28-31 maio 2017
%I Instituto Nacional de Pesquisas Espaciais (INPE)
%J São José dos Campos
%P 4406-4413
%S Anais
%1 Instituto Nacional de Pesquisas Espaciais (INPE)
%X The leaf area index (LAI) is an important parameter in the forest and interferes in various ecosystem processes. Its determination via hemispherical photography are cheap, operationally simple and reliable when the correct methods are applied. The LAI is obtained by the inversion of the gap fraction viewed in different zenith angles from a single point. The setting of the camera determines the level of contrast between the canopy and the sky and, thus, affect the capacity of the software to correct discern these two classes in the later classification step. Also, the utilization of automatic thresholding or supervised classification interferes greatly in the precision and replicability of the study. Three settings in the camera were tested, one automatic and two manual, and the two methods of classification, automatic and supervised. The study was run in a fragment of secondary Atlantic Forest in Santa Catarina, where 15 experimental points were randomly defined, in each point 3 photos were taken, one in each camera setting. After, the LAI were determined in two separate software, using automatic and supervised classification of the photos. There were a systematic underestimation of the LAI using the automatic setting of the camera, but the two manual setting returned similar results. The supervised classification overestimated the LAI in all camera settings. Based on these preliminary results, are recommended the utilization of manual mode for the photo acquisition and automatic threshold for the classification step. Further studies are needed to determine a better protocol for the photo acquisition.
%9 Processamento de imagens
%@language pt
%3 59884.pdf