Re-ranking algorithm using post-retrieval clustering for content-based image retrieval

• Published in: Information processing & management 2005-03, Vol.41 (2), p.177-194
• Main Authors: Park, Gunhan, Baek, Yunju, Lee, Heung-Kyu
• Format: Article
• Language: English
• Subjects: Algorithms; Computerized information storage and retrieval; Content analysis; Exact sciences and technology; Hierarchical clustering; Image retrieval; Information and communication sciences; Information retrieval systems. Information and document management system; Information science. Documentation; Post-retrieval clustering; Re-ranking algorithm; Sciences and techniques of general use; Searching; Similarity relationship; Studies; System design and modelling
• ISSN: 0306-4573; 1873-5371
• DOI: 10.1016/j.ipm.2003.08.002

In this paper, we propose a re-ranking algorithm using post-retrieval clustering for content-based image retrieval (CBIR). In conventional CBIR systems, it is often observed that images visually dissimilar to a query image are ranked high in retrieval results. To remedy this problem, we utilize the similarity relationship of the retrieved results via post-retrieval clustering. In the first step of our method, images are retrieved using visual features such as color histogram. Next, the retrieved images are analyzed using hierarchical agglomerative clustering methods (HACM) and the rank of the results is adjusted according to the distance of a cluster from a query. In addition, we analyze the effects of clustering methods, query-cluster similarity functions, and weighting factors in the proposed method. We conducted a number of experiments using several clustering methods and cluster parameters. Experimental results show that the proposed method achieves an improvement of retrieval effectiveness of over 10% on average in the average normalized modified retrieval rank (ANMRR) measure.