How to Cope with the Performance Gap in Content-Based Image Retrieval Systems

How to Cope with the Performance Gap in Content-Based Image Retrieval Systems

Agma J.M. Traina (University of São Paulo at São Carlos, Brazil), Caetano Traina Jr. (University of São Paulo at São Carlos, Brazil), Cristina D.A. Ciferri (University of São Paulo at São Carlos, Brazil), Marcela X. Ribeiro (University of São Paulo at São Carlos, Brazil) and Paulo M. Azevedo-Marques (University of São Paulo at Ribeirão Preto, Brazil)
DOI: 10.4018/jhisi.2009010104
OnDemand PDF Download:
$37.50

Abstract

This paper discusses the main aspects regarding the performance gap in Content-based Image Retrieval (CBIR) systems, which is an important issue regarding their acceptability. We also detail the main problems that lead to the performance gap: the use of many features to represent images, the lack of appropriate indexing structures for images and features, deficient query plans employed to execute similarity queries, and sometimes the poor quality of results obtained by the CBIR system. We present guidelines to overcome these problems by employing feature selection techniques to beat the "dimensionality curse", by using proper access methods to support fast and effective indexing and retrieval of images, by stressing the importance of using query optimization approaches and by including the user during the tuning of the CBIR system through relevance feedback techniques.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 12: 4 Issues (2017): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2016)
Volume 10: 4 Issues (2015)
Volume 9: 4 Issues (2014)
Volume 8: 4 Issues (2013)
Volume 7: 4 Issues (2012)
Volume 6: 4 Issues (2011)
Volume 5: 4 Issues (2010)
Volume 4: 4 Issues (2009)
Volume 3: 4 Issues (2008)
Volume 2: 4 Issues (2007)
Volume 1: 4 Issues (2006)
View Complete Journal Contents Listing