ABSTRACT
Content-Based Image Retrieval is a challenging problem both in terms of effectiveness and efficiency. In this paper, we present a flexible cluster-and-search approach that is able to reuse any previously proposed image descriptor as long as a suitable similarity function is provided. In the clustering step, the image data set is clustered using a hybrid divisive-agglomerative hierarchical clustering technique. The obtained clusters are organized in a tree that can be traversed efficiently using the similarity function associated with the chosen image descriptors. Our experiments have shown that we can improve search-time performance by a factor of 10 or more, at the cost of small loss in effectiveness (typically less than 15%) when compared to the state-of-the-art solutions.
- Antani, S., Long, R., Thoma, G.: Content-based image retrieval for large biomedical image archives. In: MEDINFO (2004)Google Scholar
- Bhatia, S.: Hierarchical clustering for image databases. In: Intl. Conference on Electro Information Technology, pp. 6-12 (2005)Google Scholar
- Bimbo, A.D.: Visual Information Retrieval, 1st edn. Morgan Kaufmann, San Francisco (1999)Google Scholar
- Bishop, C.: Pattern Recognition and Machine Learning, 1st edn. Springer, Heidelberg (2006) Google Scholar
- Thies, C., Malik, A., Keysers, D., et al.: Hierarchical feature clustering for CBIR in medical image databases. In: Medical Imaging, pp. 598-608 (2003)Google Scholar
- Han, J., Kamber, M.: Data Mining: Concepts and Techniques, 1st edn. Morgan Kaufmann, San Francisco (2005) Google Scholar
- Kinoshenko, D., Mashtalir, V., Yegorova, E.: Hierarchical Partitions for Content Image Retrieval from Large-Scale Database. In: Perner, P., Imiya, A. (eds.) MLDM 2005. LNCS (LNAI), vol. 3587, pp. 445-455. Springer, Heidelberg (2005) Google Scholar
- Shyu, M.-L., Chen, S.-C., Chen, M., et al.: A unified framework for image database clustering and CBIR. In: MMDBS, pp. 19-27 (2004) Google Scholar
- Pass, G., Zabih, R., Miller, J.: Comparing images using color coherence vectors. In: ACMMM (1997) Google Scholar
- Seo, J., Shneiderman, B.: Interactive Exploration of Multidimensional Microarray Data: Scatterplot Ordering, Gene Ontology Browser, and Profile Search. Phd thesis, University of Maryland, College Park (2003)Google Scholar
- Stehling, R., Nascimento, M., Falcão, A.: An adaptive and efficient clustering-based approach for CBIR in image databases. In: IDEAS, pp. 356-365 (2001) Google Scholar
- Stehling, R., Nascimento, M., Falcão, A.: A compact and efficient image retrieval approach based on border/interior classification. In: CIKM, pp. 102-109 (2002) Google Scholar
- Swain, M.J., Ballard, D.H.: Color indexing. IJCV 7(1), 11-32 (1991) Google Scholar
- Wu, W., Xiong, H., Shekhar, S. (eds.): Clustering and Information Retrieval. Kluwer, Dordrecht (2003) Google Scholar
Index Terms
- Efficient and Flexible Cluster-and-Search for CBIR
Recommendations
Energy-efficient filtering for skyline queries in cluster-based sensor networks
Display Omitted HighlightsSkyline retrieval for anti-correlated (or clustered) datasets in sensor networks is challenging.In anti-correlated and clustered datasets, many non-skyline tuples cannot be pruned by a single filtering point.We proposed a ...
An Efficient Formulation of the Improved Visual Assessment of Cluster Tendency (iVAT) Algorithm
The VAT algorithm is a visual method for determining the possible number of clusters in, or the cluster tendency of a set of objects. The improved VAT (iVAT) algorithm uses a graph-theoretic distance transform to improve the effectiveness of the VAT ...
Automatic Cluster Selection Using Index Driven Search Strategy
AI*IA '09:: Proceedings of the XIth International Conference of the Italian Association for Artificial Intelligence Reggio Emilia on Emergent Perspectives in Artificial IntelligenceClustering is the task of categorizing objects into different classes in an unsupervised way. Hierarchical clustering algorithms are usually very effective in detecting the dataset underlying structure. However, they do not create clusters, but compute ...
Comments