Image coding based on sparsified Contourlet and adjusted SPIHT

Abstract

The benefits of the directional representation are discounted by the redundancy of Contourlet transform when it is directly coded. In this paper, we propose an image compression algorithm based on sparsified Contourlet coefficients and A-SPIHT-CT (adjusted SPIHT based on Contourlet transform). The Contourlet transform is sparsified by the iterative thresholding mehod before coding. The experiments show that the coding results based on the sparsified Contourlet outperform the original Contourlet transform.