Image Retrieval Using the Double Density Wavelet Transform
Zhiyong An, Yan Liu, Zhao Feng -, Jinjiang Li · Journal of Convergence Information Technology · 2012
A novel image retrieval algorithm using the double density DWT is proposed in this paper. The double density DWT has the advantages of good directionality, nearly shift-invariance and limited redundancy. Our simulation results demonstrate that the high frequency sub-band coefficient of double density DWT can be described by the generalized Gaussian density (GGD). Then the parameters of GGD can denote the texture feature of high frequency sub-band. The local binary pattern (LBP) histogram is used to describe the texture feature of low frequency sub-band. Experiments show that the proposed algorithm using the double density DWT outperforms the SD algorithm based on the Contourlet in the natural image retrieval.