Study of neighborhood search-based fractal image encoding

Abstract

Fractal image encoding is one of the famous lossy encoding techniques ascertain high compression ratio, higher peak signal-to-noise ratio (PSNR), and good quality of encoded image. Fractal image compression uses the self-similarity property present in the natural image and similarity measure. The main drawback fractal image encoding suffers from is significant time consumption in search of appropriate domain for each range of image blocks. There have been various researches carried out to overcome the limitation of fractal encoding and to speed up the encoder. Initially, various classification and partitioning schemes were used to reduce the search space. A remarkable improvement was made by neighborhood region search strategies, which classify the image blocks on the basis of some feature vectors of image to restrict the region for best matching pair of domain and range, and also reduces the search complexity to logarithmic time. In this paper, three image block preprocessing approaches using neighborhood search method are explained in different domains and all these approaches are compared on the basis of their simulation results.