@Article{SotheAlScLiCuBeFe:2020:CoMaDe,
author = "Sothe, Camile and Almeida, Cl{\'a}udia Maria de and Schimalski,
Marcos Benedito and Liesenberg, Veraldo and Cue, Laura Elena and
Bermudez, Jos{\'e} David and Feitosa, Raul Queiroz",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {Universidade do
Estado de Santa Catarina (UDESC)} and {Universidade do Estado de
Santa Catarina (UDESC)} and {Pontif{\'{\i}}cia Universidade
Cat{\'o}lica do Rio de Janeiro (PUC-Rio)} and
{Pontif{\'{\i}}cia Universidade Cat{\'o}lica do Rio de Janeiro
(PUC-Rio)} and {Pontif{\'{\i}}cia Universidade Cat{\'o}lica do
Rio de Janeiro (PUC-Rio)}",
title = "A comparison of machine and deep learning algorithms applied to
multisource data for a subtropical forest area classification",
journal = "International Journal of Remote Sensing",
year = "2020",
volume = "41",
number = "5",
pages = "1943--1969",
note = "Setores de Atividade: Atividades dos servi{\c{c}}os de tecnologia
da informa{\c{c}}{\~a}o, Pesquisa e desenvolvimento
cient{\'{\i}}fico, Produ{\c{c}}{\~a}o Florestal.",
keywords = "forest succession stages, endangered tree species, convolutional
neural networks, ensemble methods, light detection and ranging,
multispectral data.",
abstract = "This work explores the integration of airborne Light Detection and
Ranging (LiDAR) data and WorldView-2 (WV2) images to classify the
land cover of a subtropical forest area in Southern Brazil.
Different deep and machine learning methods were used: one based
on convolutional neural network (CNN) and three ensemble methods.
We adopted both pixel- (in the case of CNN) and object-based
approaches. The results demonstrated that the integration of LiDAR
and WV2 data led to a significant increase (7% to 16%) in
accuracies for all classifiers, with kappa coefficient (\κ)
ranging from 0.74 for the random forest (RF) classifier associated
with the WV2 dataset, to 0.92 for the forest by penalizing
attributes (FPA) with the full (LiDAR + WV2) dataset. Using the
WV2 dataset solely, the best \κ was 0.81 with CNN
classifier, while for the LiDAR dataset, the best \κ was 0.8
with the rotation forest (RotF) algorithm. The use of LiDAR data
was especially useful for the discrimination of vegetation classes
because of the different height properties among them. In its
turn, the WV2 data provided better performance for classes with
less structure variation, such as field and bare soil. All the
classification algorithms had a nearly similar performance: the
results vary slightly according to the dataset used and none of
the methods achieved the best accuracy for all classes. It was
noticed that both datasets (WV2 and LiDAR) even when applied alone
achieved good results with deep and machine learning methods.
However, the advantages of integrating active and passive sensors
were evident. All these methods provided promising results for
land cover classification experiments of the study area in this
work.",
doi = "10.1080/01431161.2019.1681600",
url = "http://dx.doi.org/10.1080/01431161.2019.1681600",
issn = "0143-1161",
label = "lattes: 1861914973833506 2 S{\"o}theAlScLiRoBeFe:2019:CoMaDe",
language = "en",
targetfile = "sothe_comparison.pdf",
url = "https://www.tandfonline.com/doi/full/10.1080/01431161.2019.1681600",
urlaccessdate = "28 abr. 2024"
}