Adaptive image retrieval based on generalized Gaussian model and LBP

This paper proposes an adaptive image retrieval method via spatial-frequency mixed features (SFMF). The SFMF can describe spatial and frequency information of image simultaneously. More specifically, spatial feature is local binary pattern (LBP) histogram extracted from image. Frequency features are described as the generalized Gaussian density (GGD) of Contourlet transform detail coefficients and LBP histogram of approximation coefficients. Further, we use closed-loop feedback to adjust weighting factor adoptively for image retrieval. Experiments show that average recall rate of this method is 12.08%, 10.23% higher than frequency domain method and LBP respectively.