Abstract—The objective of Content-Based Image Retrieval (CBIR) methods is essentially to extract, from large (image) databases, a specified number of images similar in visual and semantic content to a so- called query image. To bridge the semantic gap that exists between the representation of an image by low-level features (namely, colour, shape, texture) and its high-level semantic content as perceived by humans, CBIR systems typically make use of the relevance feedback (RF) mechanism. RF iteratively incorporates user-given inputs regarding the relevance of retrieved images, to improve retrieval efficiency. One approach is to vary the weights of the features dynamically via feature reweighting. In this work, a novel approach has been proposed for improving the retrieval accuracy of CBIR system which incorporates RF based on feature reweighting using discriminant analysis. Results of a number of experiments have been presented to illustrate the significant improvement is retrieval accuracy with the proposed approach.
Index Terms—Content-based image retrieval, Discriminant Analysis, Feature reweighting, Relevance feedback.
Smarajit Bose and Amita Pal are with the Interdisciplinary Statistical Research Unit, Applied Statistics Division, Indian Statistical Institute, Kolkata, India (e-mail: {smarajit,pamita}@isical.ac.in).
Disha Chakrabarti is with Tata Consultancy Services, Kolkata, India (e-mail: disha.chakrabarti@gmail.com).
Taranga Mukherjee is with the Department of Statistics, University of Calcutta, Kolkata, India (e-mail: tm.custat@gmail.com).
Cite: Smarajit Bose, Amita Pal, Disha Chakrabarti, and Taranga Mukherjee, "Improved Content-Based Image Retrieval via Discriminant Analysis," International Journal of Machine Learning and Computing vol. 7, no. 3, pp. 44-48, 2017.