@Article{NegriDutrSantLu:2016:ExReMe,
author = "Negri, Rog{\'e}rio G. and Dutra, Luciano Vieira and Sant'Anna,
Sidnei Jo{\~a}o Siqueira and Lu, D.",
affiliation = "{Universidade Estadual Paulista (UNESP)} and {Instituto Nacional
de Pesquisas Espaciais (INPE)} and {Instituto Nacional de
Pesquisas Espaciais (INPE)} and {Michigan State University}",
title = "Examining region-based methods for land cover classification using
stochastic distances",
journal = "International Journal of Remote Sensing",
year = "2016",
volume = "37",
number = "8",
pages = "1902--1921",
month = "Apr.",
keywords = "Graph theory, Pixels, Radar imaging, Remote sensing, Stochastic
systems, Support vector machines, Synthetic aperture radar.",
abstract = "A recent alternative to standard pixel-based classification of
remote-sensing data is region-based classification, which has
proved to be particularly useful when analysing high-resolution
imagery of complex environments, such as urban areas, or when
addressing noisy data, such as synthetic aperture radar (SAR)
images. First, following certain criteria, the imagery is
decomposed into homogeneous regions, and then each region is
classified into a class of interest. The usual method for
region-based classification involves using stochastic distances,
which measure the distances between the pixel distributions inside
an unknown region and the representative distributions of each
class. The class, which is at the minimum distance from the
unknown region distribution, is assigned to the region and this
procedure is termed stochastic minimum distance classification
(SMDC). This study reports the use of methods derived from the
original SMDC, Support Vector Machine (SVM), and graph theory,
with the objective of identifying the most robust and accurate
classification methods. The equivalent pixel-based versions of
region-based analysed methods were included for comparison. A case
study near the Tapaj{\'o}s National Forest, in Par{\'a} state,
Brazil, was investigated using ALOS PALSAR data. This study showed
that methods based on the nearest neighbour, derived from SMDC,
and SVM, with a specific kernel function, are more accurate and
robust than the other analysed methods for region-based
classification. Furthermore, pixel-based methods are not indicated
to perform the classification of images with a strong presence of
noise, such as SAR images.",
doi = "10.1080/01431161.2016.1165883",
url = "http://dx.doi.org/10.1080/01431161.2016.1165883",
issn = "0143-1161",
language = "en",
urlaccessdate = "02 maio 2024"
}