Further Results on Texture Discrimination Based upon an Assumed Stochastic Texture Model.
Acie L. Vickers, James W. Modestino · 1981
Further results of a new approach to texture discrimination are described. The approach is based upon an assumed stochastic model for texture in imagery and is an approximation to the statistically optimum maximum likelihood classifier. A method for estimating the structural parameters of the assumed stochastic texture model for each known texture class is described, and experimental results are provided which demonstrate the efficacy of this approach with simulated and real world texture data. (Author)