Information-theoretic selection of high-dimensional spectral features for structural recognition

Please use this identifier to cite or link to this item: http://hdl.handle.net/10045/52221
Información del item - Informació de l'item - Item information
Title: Information-theoretic selection of high-dimensional spectral features for structural recognition
Authors: Bonev, Boyan | Escolano, Francisco | Giorgi, Daniela | Biasotti, Silvia
Research Group/s: Laboratorio de Investigación en Visión Móvil (MVRLab)
Center, Department or Service: Universidad de Alicante. Departamento de Ciencia de la Computación e Inteligencia Artificial
Keywords: Feature selection | Pattern classification | Information theory | Mutual information | Entropy | Structure | Spectral features
Knowledge Area: Ciencia de la Computación e Inteligencia Artificial
Issue Date: Mar-2013
Publisher: Elsevier
Citation: Computer Vision and Image Understanding. 2013, 117(3): 214-228. doi:10.1016/j.cviu.2012.11.007
Abstract: Pattern recognition methods often deal with samples consisting of thousands of features. Therefore, the reduction of their dimensionality becomes crucial to make the data sets tractable. Feature selection techniques remove the irrelevant and noisy features and select a subset of features which describe better the samples and produce a better classification performance. In this paper, we propose a novel feature selection method for supervised classification within an information-theoretic framework. Mutual information is exploited for measuring the statistical relation between a subset of features and the class labels of the samples. Traditionally it has been measured for ranking single features; however, in most data sets the features are not independent and their combination provides much more information about the class than the sum of their individual prediction power. We analyze the use of different estimation methods which bypass the density estimation and estimate entropy and mutual information directly from the set of samples. These methods allow us to efficiently evaluate multivariate sets of thousands of features. Within this framework we experiment with spectral graph features extracted from 3D shapes. Most of the existing graph classification techniques rely on the graph attributes. We use unattributed graphs to show what is the contribution of each spectral feature to graph classification. Apart from succeeding to classify graphs from shapes relying only on their structure, we test to what extent the set of selected spectral features are robust to perturbations of the dataset.
URI: http://hdl.handle.net/10045/52221
ISSN: 1077-3142 (Print) | 1090-235X (Online)
DOI: 10.1016/j.cviu.2012.11.007
Language: eng
Type: info:eu-repo/semantics/article
Rights: © 2012 Elsevier Inc.
Peer Review: si
Publisher version: http://dx.doi.org/10.1016/j.cviu.2012.11.007
Appears in Collections:INV - MVRLab - Artículos de Revistas

Files in This Item:
Files in This Item:
File Description SizeFormat 
Thumbnail2013_Bonev_etal_CVIU_final.pdfVersión final (acceso restringido)2,44 MBAdobe PDFOpen    Request a copy


Items in RUA are protected by copyright, with all rights reserved, unless otherwise indicated.