Properties of series feature aggregation schemes

Zhang, Jun and Ye, Lei 2010, Properties of series feature aggregation schemes, World review of science, technology and sustainable development, vol. 7, no. 1-2, pp. 100-115, doi: 10.1504/WRSTSD.2010.032347.

Attached Files
Name Description MIMEType Size Downloads

Title Properties of series feature aggregation schemes
Author(s) Zhang, JunORCID iD for Zhang, Jun
Ye, Lei
Journal name World review of science, technology and sustainable development
Volume number 7
Issue number 1-2
Start page 100
End page 115
Total pages 16
Publisher Inderscience Publishers
Place of publication Geneva, Switzerland
Publication date 2010-03
ISSN 1741-2242
Keyword(s) CBIR
content-based image retrieval
feature fusion
series feature aggregation
threshold estimation
Summary Feature aggregation is a critical technique in content-based image retrieval (CBIR) that combines multiple feature distances to obtain image dissimilarity. Conventional parallel feature aggregation (PFA) schemes failed to effectively filter out the irrelevant images using individual visual features before ranking images in collection. Series feature aggregation (SFA) is a new scheme that aims to address this problem. This paper investigates three important properties of SFA that are significant for design of systems. They reveal the irrelevance of feature order and the convertibility of SFA and PFA as well as the superior performance of SFA. Furthermore, based on Gaussian kernel density estimator, the authors propose a new method to estimate the visual threshold, which is the key parameter of SFA. Experiments, conducted with IAPR TC-12 benchmark image collection (ImageCLEF2006) that contains over 20,000 photographic images and defined queries, have shown that SFA can outperform conventional PFA schemes.
Language eng
DOI 10.1504/WRSTSD.2010.032347
Field of Research 080704 Information Retrieval and Web Search
080109 Pattern Recognition and Data Mining
Socio Economic Objective 890301 Electronic Information Storage and Retrieval Services
HERDC Research category C1.1 Refereed article in a scholarly journal
Copyright notice ©2010, Inderscience Enterprises Ltd.
Persistent URL

Document type: Journal Article
Collection: School of Information Technology
Connect to link resolver
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 1 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 238 Abstract Views, 5 File Downloads  -  Detailed Statistics
Created: Wed, 26 Oct 2011, 12:48:12 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact