Entropy-Based Measures for Clustering and SOM Topology Preservation Applied to Content-Based Image Indexing and Retrieval

Authors:
Markus Koskela;Jorma Laaksonen;Erkki Oja
Affiliations:
Helsinki University of Technology, Finland;Helsinki University of Technology, Finland;Helsinki University of Technology, Finland
Venue:
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 2 - Volume 02
Year:
2004

Citing 0
Cited 7

Dynamic training using multistage clustering for face recognition

Pattern Recognition
An automated approach for the optimization of pixel-based visualizations

Information Visualization
DETS: a meta heuristic approach for document retrieval

International Journal of Knowledge Engineering and Soft Data Paradigms
Entropy-based iterative face classification

BioID'11 Proceedings of the COST 2101 European conference on Biometrics and ID management
Topographic measure based on external criteria for self-organizing map

WSOM'11 Proceedings of the 8th international conference on Advances in self-organizing maps
Partial relevance in interactive facial image retrieval

ICAPR'05 Proceedings of the Third international conference on Pattern Recognition and Image Analysis - Volume Part II
Visual query processing for efficient image retrieval using a SOM-based filter-refinement scheme

Information Sciences: an International Journal

Quantified Score

Hi-index	0.01

Visualization

Abstract

Content-based image retrieval (CBIR) addresses the problem of finding images relevant to the usersý information needs, based principally on low-level visual features for which automatic extraction methods are available. For the development of CBIR applications, an important issue is to have efficient and objective performance assessment methods for different features and techniques. In this paper, we study the efficiency of clustering methods for image indexing with entropy-based measures. Furthermore, the Self-Organizing Map (SOM) as an indexing method is discussed further and an analysis method which takes into account also the spatial configuration of the data on the SOMis presented. The proposed methods enable computationally light measurement of indexing and retrieval performance for individual image features.