Combined sparse representation based on curvelet transform and local DCT for multi-layered image compression

In this paper, we propose a new multi-layered representation technique for image compression, which combine curvelet transform and local DCT in order to benefit from the advantages of each. Curvelet transform is one of the recently developed multiscale transform, which possess directional features and provides optimally sparse representation of objects with edges, but not for the textured feature. We exploit morphological component analysis (MCA) method to separate the image into two layers: piecewise smooth layer and textured structure layer, respectively associated to curvelet transform and local DCT. Each layer is encoded independently with a different transform at a different bitrate. Experiment results show that the proposed multi-layered image coding technique outperforms single curvelet transform with SPIHT in both PSNR and visual quality, with the improvement up to 0.71 dB on images with rich texture and edges.