Title :
A class of multiresolution stochastic models generating self-affine images
Author :
Wohlberg, Brendt ; De Jager, Gerhard
Author_Institution :
Lab. of Digital Image Process., Cape Town Univ., Rondebosch, South Africa
fDate :
6/1/1999 12:00:00 AM
Abstract :
Fractal image compression is based on the rather poorly motivated assumption that “natural” images exhibit significant affine self-similarity. The accuracy of this assumption is evaluated by a comparison between the statistics of natural images and those of a multiresolution stochastic model designed to generate images exhibiting affine self-similarity as assumed by fractal coding techniques. These comparisons suggest that self-affinity does not represent a particularly accurate characterization of image statistics
Keywords :
data compression; fractals; higher order statistics; image coding; stochastic processes; wavelet transforms; affine self-similarity; fractal coding techniques; fractal image compression; higher order statistics; image statistics characterization; multiresolution stochastic models; natural images; self-affine images; wavelet transforms; Adaptive filters; Fractals; Higher order statistics; Image coding; Image generation; Image resolution; Signal processing algorithms; Signal resolution; Stochastic processes; Wavelet transforms;
Journal_Title :
Signal Processing, IEEE Transactions on
