Traditional wavelet-based image coders use dyadic multiresolution decompositions, resulting in spatial resolutions that are also dyadic in scale, i.e., powers of 1/2 the original resolution. In this paper, we increase the set of decodable spatial resolutions beyond dyadic scales by introducing a two-dimensional wavelet decomposition made of combining wavelet transforms of non-integer dilation factors. We describe how the proposed wavelet decomposition can produce spatial resolutions related to the original resolution by dyadic and non-dyadic ratios, and present how the resulting subband structure can be made compatible with zerotree-based subband coders. © Springer-Verlag Berlin Heidelberg 2010.
CITATION STYLE
Petngang, F., Dansereau, R. M., & Joslin, C. (2010). Resolution scalable image coding with dyadic complementary rational wavelet transforms. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6134 LNCS, pp. 28–35). https://doi.org/10.1007/978-3-642-13681-8_4
Mendeley helps you to discover research relevant for your work.