Authors: I. Felci Rajam; S. Valli
Addresses: Department of M.C.A., St. Joseph's College of Engineering, Jeppiaar Nagar, Old Mamallapuram Road, Chennai 600119, India; Department of C.S.E., Anna University, Chennai 600025, Tamil Nadu, India. ' Department of CSE, College of Engineering, Anna University, Sardar Patel Road, Guindy, Chennai 600025, Tamil Nadu, India
Abstract: An image retrieval system, which reduces the semantic gap between the low level image features and the high level semantic concepts, using the semantic cluster matrix (SCM) and adaptive learning during testing, is proposed. This mechanism retrieves semantically relevant images for both trained and untrained image categories. The SCM groups the new categories into semantic clusters, records the cluster's semantic information, gets the relevance feedback (RF) from the user, and records this in the SCM. Thus, the SCM adaptively learns about the new categories during the testing time, and is able to retrieve semantically relevant images for untrained categories also. Experiments were conducted using the Caltech image dataset consisting of 101 categories of images. The obtained results demonstrate that the proposed approach achieves good performance in terms of retrieval accuracy.
Keywords: semantic clustering; semantic template; semantic threshold; affinity matrix; feature extraction; semantic cluster matrix; SCM; support vector machine; SVM; binary decision tree; BDT; relevance feedback; adaptive learning; region-based image retrieval; RBIR; semantically relevant images.
International Journal of Computational Science and Engineering, 2012 Vol.7 No.3, pp.239 - 252
Received: 02 Oct 2011
Accepted: 26 Feb 2012
Published online: 24 Jul 2012 *