Region-oriented compression of color images using fuzzy inference and shape adaptive DCT

Abstract

A new region-based compression scheme for color images is proposed. The segmentation method belongs to the split and merge category. Splitting is carried out using the watershed transform. In the merging stage, a fuzzy color preserving rule-based system is introduced to express the dissimilarity between two regions. The compression part is based on the shape adaptive DCT with /spl Delta/DC correction method. The quantization matrices have been designed according to the properties of the used transform. This method yields promising results, mostly attributed to the efficiency of the segmentation and particularly the fuzzy dissimilarity estimation.