@InProceedings{ArvorSaiDupAndDur:2013:IdOpCl,
author = "Arvor, Damien and Saint-Geours, Nathalie and Dupuy, St{\'e}phane
and Andr{\'e}s, Samuel and Durieux, Laurent",
title = "Identifying optimal classification rules for geographic
object-based image analysis",
booktitle = "Anais...",
year = "2013",
editor = "Epiphanio, Jos{\'e} Carlos Neves and Galv{\~a}o, L{\^e}nio
Soares",
pages = "2290--2297",
organization = "Simp{\'o}sio Brasileiro de Sensoriamento Remoto, 16. (SBSR)",
publisher = "Instituto Nacional de Pesquisas Espaciais (INPE)",
address = "S{\~a}o Jos{\'e} dos Campos",
abstract = "In Geographic Object-based Image Analysis (GEOBIA), remote sensing
experts benefit from a large spectrum of characteristics to
interpret images (spectral information, texture, geometry, spatial
relations, etc). However, the quality of a classification is not
always increased by considering a higher number of features. The
experts are then used to define classification rules based on a
laborious {"}trial-and-error{"} process. In this paper, we test a
methodology to automatically determine an optimal subset of
features for discriminating features. This method assumes that a
reference land cover map (or at least training samples) is
available. Two approaches were considered: a rule-based approach
and a Support Vector Machine approach. For each approach, the
method consists in ranking the features according to their
potential for discriminating two classes. This task was performed
thanks to the Jeffries-Matusita distance and Support Vector
Machine-Ranking Feature Extraction (SVM-RFE) algorithm. Then, it
consists in training and validating a classification algorithm
(rule-based and SVM), with an increasing number of features: first
only the best-ranked feature is included in the classifier, then
the two best-ranked features, etc., until all the N features are
included. The objective is to analyze how the quality of the
classification evolves according to the numbers of features used.
The optimal subset of features is finally determined through the
analysis of the Akaike information criterion. The methodology was
tested on two classes (urban an non urban areas) on a Spot5 image
regarding a study area located in the La R{\'e}union island.",
conference-location = "Foz do Igua{\c{c}}u",
conference-year = "13-18 abr. 2013",
isbn = "{978-85-17-00066-9 (Internet)} and {978-85-17-00065-2 (DVD)}",
label = "1605",
language = "en",
organisation = "Instituto Nacional de Pesquisas Espaciais (INPE)",
ibi = "3ERPFQRTRW34M/3E7GMB5",
url = "http://urlib.net/ibi/3ERPFQRTRW34M/3E7GMB5",
targetfile = "p1605.pdf",
type = "Classifica{\c{c}}{\~a}o e Minera{\c{c}}{\~a}o de Dados",
urlaccessdate = "12 maio 2024"
}