Title : ( A NEW IMPROVED WAVELET CORRELOGRAM METHOD FOR IMAGE INDEXING AND RETRIEVAL )
Authors: Mahdi Saadatmand , H. Abrishami Moghaddam ,
Full TextAbstract
In this paper, a new algorithm called wavelet correlogram-CCV (W3C) for content-based image indexing and retrieval is presented. W3C is an extension of wavelet correlogram that utilizes color coherence vectors (CCV) to consider texture information in addition to the shape for gray-level images. Additionally, a new distance measure that is optimized by genetic algorithms is proposed. The retrieval results obtained by the new method on a 1000 non-medical image database demonstrate significant improvements in recall and rank compared to wavelet correlogram and SIMPLIcity methods. Also, simulations on a 1030 medical image database show considerable good performance.
Keywords
, Content Based Image Indexing and Retrieval, Wavelet Correlogram, Color Coherence Vector, Genetic Algorithms, Medical Imagebases
@inproceedings{paperid:1030347,
author = {Saadatmand, Mahdi and H. Abrishami Moghaddam},
title = {A NEW IMPROVED WAVELET CORRELOGRAM METHOD FOR IMAGE INDEXING AND RETRIEVAL},
booktitle = {چهارمین همایش سراسری مهندسی کامپیوتر},
year = {2006},
location = {تهران, IRAN},
keywords = {Content Based Image Indexing and Retrieval; Wavelet Correlogram; Color Coherence Vector; Genetic Algorithms; Medical Imagebases},
}
%0 Conference Proceedings
%T A NEW IMPROVED WAVELET CORRELOGRAM METHOD FOR IMAGE INDEXING AND RETRIEVAL
%A Saadatmand, Mahdi
%A H. Abrishami Moghaddam
%J چهارمین همایش سراسری مهندسی کامپیوتر
%D 2006
