Wavelet image compression based on significance clustering and rate distortion optimization

This paper presents a new efficient wavelet-based image compression algorithm. Morphological dilation is applied to extract the clustered significant coefficients in each subband resulting in the partitioning of each subband into significance clusters and insignificance space. With this partitioning, the rate distortion is optimized in the proposed algorithm by encoding the significance clusters in all subbands first. When encoding the insignificance space, the zerotree is discovered to be not very efficient for representing zeros across scales for texture images, and a more efficient method is proposed. Experimental results show that the performance of the proposed algorithm compares favorably with the most efficient wavelet-based image compression algorithm published so far.