Set Partitioning in Hierarchical Frequency Bands (SPHFB)

Abstract

A novel algorithm for very low bit rate based on hierarchical partition of subbands in the wavelet domain is proposed. The algorithm uses the set partitioning technique to sort the transformed coefficients. The threshold of each subband is calculated and the subbands scanning sequence is determined by the magnitude of the thresholds which establish a hierarchical scanning not only for the set of coefficients with large magnitude, but also for the subbands. Results show that SPHFB provides good image quality for very low bit rates.