Accelerating of Image Retrieval in CBIR System with Relevance Feedback. (26th June 2007)

• Record Type: Journal Article
• Authors: Zajić Zajić, Goran Goran; Kojić Kojić, Nenad Nenad; Radosavljević Radosavljević, Vladan Vladan; Rudinac Rudinac, Maja Maja; Rudinac Rudinac, Stevan Stevan; Reljin Reljin, Nikola Nikola; Reljin Reljin, Irini Irini; Reljin Reljin, Branimir Branimir
• Other Names: Izquierdo Izquierdo Ebroul Ebroul Academic Editor.
• Is Part Of: EURASIP journal on advances in signal processing. Volume 2007(2007)
• Publication Date: 2007-06-26
• Subjects: Signal processing -- Periodicals; Traitement du signal; Signal processing; Periodicals; 621.3822
• DOI: 10.1155/2007/62678
• Languages: English
• ISSNs: 1687-6172

Abstract

Abstract : Content-based image retrieval (CBIR) system with relevance feedback, which uses the algorithm for feature-vector (FV) dimension reduction, is described. Feature-vector reduction (FVR) exploits the clustering of FV components for a given query. Clustering is based on the comparison of magnitudes of FV components of a query. Instead of all FV components describing color, line directions, and texture, only their representative members describing FV clusters are used for retrieval. In this way, the "curse of dimensionality" is bypassed since redundant components of a query FV are rejected. It was shown that about one tenth of total FV components (i.e., the reduction of 90%) is sufficient for retrieval, without significant degradation of accuracy. Consequently, the retrieving process is accelerated. Moreover, even better balancing between color and line/texture features is obtained. The efficiency of FVR CBIR system was tested over TRECVid 2006 and Corel 60 K datasets.