IMAGE THRESHOLDING BASED ON ALI-SILVEY DISTANCE MEASURES

A relative entropy based approach to image thresholding was proposed r ecently. It was demonstrated that this method was successful for image thresholding. Relative entropy is a member of the class of Ali-Silvey distance measures. In this paper we generalize the relative entropy b ased approach and present image thresholding algorithms based on the c lass of Ali-Silvey distance measures. A number of members of this clas s are selected and used for implementation in image thresholding algor ithms. Performance is evaluated by applying these algorithms to severa l images and comparing them to a few histogram based thresholding meth ods. (C) 1997 Pattern Recognition Society.