A smart color image retrieval method based on multiresolution features

This paper proposes a novel technique of color image retrieval based on multi-resolution image features and similarity measures, which extracts the color and texture features at optimum level of multi-resolution pyramid image. Here, Wavelet technique is employed to derive a multi-resolution pyramid image. Such extraction at an optimum level helps in formation of a feature vector. The rotation invariant based Bhattacharyya measure (BM) and orthogonal Cosine distance method are employed on the feature vectors of the query and target images for finding a similar image. The proposed method is conceptually simple, memory efficient, and suitable for fast response requirements, since the features extracted at optimum level image contain only a few dominant wavelet coefficients. The efficiency of the proposed feature vector is experimented with standard Vistex and Corel image databases. The proposed system compares with other recently developed methods such as orthogonal polynomial model, Multi-resolution with BDIP(Block Difference of Inverse Probabilities)-BVLC(Bock Variation of Local Correlation coefficients) and Wavelet moment methods.