A Fractal Image Coding Scheme Based on Norm

Authors: Atilla BAŞKURT, Kamel BELLOULATA, Gérard GIMENEZ

Abstract: In this paper we discuss the metric used for the search of self-similarity at different scales in fractal coding of images. It is well known that the quality of decoded images depends directly on the efficiency of the measurement of the self-similarity between objects of the original image. We present two coding schemes based on L_2 and L_{infinite} metrics. The L_2 metric, widely used in fractal coding algorithms, measures a global distortion. On the contrary, L_{infinite} gives the maximum grey level error between the pixels of two objects and enables this local error to be controlled. The performance of these schemes is compared with various standard images in terms of PSNR and maximum local distortion. L_{infinite} metric speeds up the coding process and avoids local distortion peaks. L_2 provides a better perceptual quality of the decoded images.

Keywords: image coding, fractal coding, L_2 metric, L_{infinite} metric.