A novel image retrieval algorithm using wavelet packet histogram techniques

This paper proposes a novel image retrieval algorithm based on wavelet packet histogram techniques. Firstly, we decompose an image with a family of real orthonormal wavelet bases and compute the energy using wavelet packet coefficients. Secondly, we select few number of most dominant energy channels for thresholding and non-linear filtering. Finally, we compute the wavelet packet histogram as features and employ histogram intersection distance to retrieve queried image from image databases, the proposed method employs a smaller feature space and involves as less computation cost in the feature extraction. The experimental results show that these techniques can archive higher performance in the image retrieval