Contourlet based image compression using controlled modification of coefficients

A new compression algorithm is proposed in this paper which uses the contourlet transform. Unlike contourlet-based non-linear approximation (NLA) compression algorithms, the proposed algorithm modifies the coefficients in a controlled manner. The modification is performed so that the difference between a modified coefficient and its original value is within a certain range. To achieve higher compression, the modifications are performed with the goal of minimizing the entropy of the coefficients. The implementation results show that our algorithm produces images with higher PSNRs, for similar bit-rate conditions, as compared to NLA compression algorithms. Furthermore, the visual quality of the images produced by our algorithm is higher than the mentioned NLA algorithms. The implementation results also show the superiority of our algorithm over WBCT algorithm which is based on the joint application of wavelet and contourlet transforms.