A low complexity wavelet image coder based on significance tree splitting

In this paper, a low complexity wavelet image coding algorithm based on significance tree splitting will be presented. In the proposed coding algorithm, we build significance trees corresponding to each spatial quadtree in wavelet domain. The splitting of significance tree is used to guide the encoding of the higher frequency subbands by indicating which of their coefficient bitplanes require encoding and which do not. The coding algorithm exploits both interband clustering of significant coefficients and intraband dependencies in significant fields by using the efficient quadtree decomposition model. Moreover, the coding algorithm has low complexity and requires low memory. Experimental results for some standard images show that proposed algorithm provides competitive compression results.