SUPERVISED TEXTURE CLASSIFICATION USING A PROBABILISTIC NEURAL-NETWORK AND CONSTRAINT SATISFACTION MODEL

In this paper, the texture classification problem is projected as a co nstraint satisfaction problem. The focus is on the use of a probabilis tic neural network (PNN) for representing the distribution of feature vectors of each texture class in order to generate a feature-label int eraction constraint. This distribution of features for each class is a ssumed as a Gaussian mixture model. The feature-label interactions and a set of label-label interactions are represented on a constraint sat isfaction neural network. A stochastic relaxation strategy is used to obtain an optimal classification of textures in an image. The advantag e of this approach is that all classes in an image are determined simu ltaneously, similar to human perception of textures in an image.