According to the collage theorem, the encoding distortion for fractal image compression is directly related to the metric used in the encoding process. In this paper, we introduce a perceptually meaningful distortion measure based on the human visual system's nonlinear response to luminance and the visual masking effects. Blackwell's psychophysical raw data on contrast threshold are first interpolated as a function of background luminance and visual angle, and are then used as an error upper bound for perceptually lossless image compression. For a variety of images, experimental results show that the algorithm produces a compression ratio of 8:1 to 10:1 without introducing visual artifacts.

Date Published: 27 February 1996
PDF: 6 pages
Proc. SPIE 2727, Visual Communications and Image Processing '96, (27 February 1996);

Published in SPIE Proceedings Vol. 2727:
Visual Communications and Image Processing '96
Rashid Ansari; Mark J. T. Smith, Editor(s)