Content based image retrieval based on BEMD: Use of curvelet transform coefficients distribution and Gabor wavelets

In this paper, we present a new method for content based images retrieval. We propose to characterize images without extracting local features, by using global information extracted from the image Bidimensional Empirical Mode Decomposition (BEMD). This method decomposes image into a set of functions named Bidimensional Intrinsic Mode Functions (BIMF) and a residue. Image feature based on curvelet transform and Gabor wavelet, are then calculated. Our system was tested on Brodatz database. Retrieval efficiency and accuracy combining image decomposition model, and both curvelet method and Gabor wavelet based features are found to be superior to other existing methods.