Applications of stochastic texture field models to image processing

Abstract

Stochastic models for the generation of two-dimensional texture fields have been used in studies of human visual texture perception. Such models also find application in image processing for the development of techniques for synthesizing and analyzing texture. This paper presents a survey of various stochastic texture models that have been developed. Results of visual texture discrimination experiments are also provided. Finally, examples are given of the application of these models to texture synthesis and texture analysis.