PERFORMANCE EVALUATION OF IMAGE QUALITY BASED ON FRACTAL IMAGE COMPRESSION

Fractal techniques for image compression haverecently attracted a great deal of attention. Fractalimage compression is a relatively recenttechnique based on the representation of animage by a contractive transform, on the space ofimages, for which the fixed point is close to theoriginal image. This broad principle encompassesa very wide variety of coding schemes, many ofwhich have been explored in the rapidly growingbody of published research.Unfortunately, littlein the way of practical algorithms or techniqueshas been published. Here present a technique forimage compression that is based on a very simpletype of iterative fractal. In our algorithm awavelet transform (quadrature mirror filterpyramid) is used to decompose an image intobands containing information from differentscales (spatial frequencies) and orientations. Theconditional probabilities between these differentscale bands are then determined, and used as thebasis for a predictive coder.We undertake a study of the performance offractal image compression. This paper focusesimportant features of compression of still images,including the extent to which the quality of imageis degraded by the process of compression anddecompression.The numerical experiment is doneby considering various types of images and byapplying fractal Image compression to compressan image. It was found that fractal yields betterresult as compared to other compressiontechniques. It provide better peak signal to noiseratio as compare to other techniques, but it takehigher encoding time.The numerical results arecalculated in Matlab.