Coherence vector based on wavelet coefficients for image retrieval

A new algorithm for content-based image indexing and retrieval is presented in this paper. The proposed method combines wavelet transform and color coherence vector. Wavelet coefficients of image are computed first using wavelet transform such as Daubechies wavelets. A quantization step is then applied before computing coherence vector of the wavelet coefficients. Finally, index vectors are constructed using these wavelet coherence vectors. The retrieval results obtained by the method on a 500 image database show the proposed algorithm achieves a significant improvement in effectiveness and efficiency compared to the indexing and retrieval methods based on histogram or color coherence vector of image.