@Article{BragaPDFTACSW:2020:TrCrDe,
author = "Braga, Jos{\'e} Renato Garcia and Peripato, Vin{\'{\i}}cius
Borges Pereira and Dalagnol da Silva, Ricardo and Ferreira,
Matheus P. and Tarabalka, Yuliya and Arag{\~a}o, Luiz Eduardo
Oliveira e Cruz de and Campos Velho, Haroldo Fraga de and
Shiguemori, Elcio H. and Wagner, Fabien Hubert",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {Instituto Nacional de
Pesquisas Espaciais (INPE)} and {Instituto Militar de Engenharia
(IME)} and {Inria Sophia Antipolis} and {Instituto Nacional de
Pesquisas Espaciais (INPE)} and {Instituto Nacional de Pesquisas
Espaciais (INPE)} and {Instituto de Estudos Avan{\c{c}}ado
(IEAv)} and {Instituto Nacional de Pesquisas Espaciais (INPE)}",
title = "Tree crown delineation algorithm based on a convolutional neural
network",
journal = "Remote Sensing",
year = "2020",
volume = "12",
number = "8",
pages = "e1288",
month = "Apr.",
keywords = "tree crown delineation, tropical forests, optical satellite
images, deep learning.",
abstract = "Tropical forests concentrate the largest diversity of species on
the planet and play a key role in maintaining environmental
processes. Due to the importance of those forests, there is
growing interest in mapping their components and getting
information at an individual tree level to conduct reliable
satellite-based forest inventory for biomass and species
distribution qualification. Individual tree crown information
could be manually gathered from high resolution satellite images;
however, to achieve this task at large-scale, an algorithm to
identify and delineate each tree crown individually, with high
accuracy, is a prerequisite. In this study, we propose the
application of a convolutional neural networkMask R-CNN
algorithmto perform the tree crown detection and delineation. The
algorithm uses very high-resolution satellite images from tropical
forests. The results obtained are promisingthe Recall, Precision,
and F1 score values obtained were were 0.81, 0.91, and 0.86,
respectively. In the study site, the total of tree crowns
delineated was 59, 062. These results suggest that this algorithm
can be used to assist the planning and conduction of forest
inventories. As the algorithm is based on a Deep Learning
approach, it can be systematically trained and used for other
regions.",
doi = "10.3390/RS12081288",
url = "http://dx.doi.org/10.3390/RS12081288",
issn = "2072-4292",
language = "en",
targetfile = "remotesensing-12-01288-v3.pdf",
urlaccessdate = "2024, Apr. 18"
}