In this paper, we propose a new content-based image retrieval system based on lifting wavelet filters. Lifting wavelet filters include controllable free parameters. A new similarity measure of images is introduced using the free parameters computed from the high frequency components of images. Using this measure, matching of query image with template images is done. Next, the query image is compared with the subimages lying under the category of template image similar to the query image using their histograms. A clustering method for subimages is also proposed. Finally, simulation results are given.
|Title of host publication||Recent Advances in Circuits, Systems and Signal Processing|
|Publisher||World Scientific and Engineering Academy and Society|
|Number of pages||6|
|Publication status||Published - 2002|
All Science Journal Classification (ASJC) codes