Relaxation operators

Abstract

AbstractThis paper discusses the correction (relaxation) operator in the stochastic relaxation model. The purpose is to arrive at a more useful relaxation operator by examining the relaxation process. Numerous studies have been made on the relaxation method in relation to the image processing and image interpretation. However, only a few studies have been made on the relaxation operator itself, which plays an important role in the relaxation method. This paper defines first the relaxation operator in general in the stochastic relaxation model. A concept is then introduced whereby the relaxation operator is adaptive, and the necessary and sufficient condition is shown for the operator to be adaptive. By constructing the actual relaxation operator satisfying the condition, it is shown that the relaxation operator proposed by Rosenfeld can also be derived as a special case of the operator satisfying the condition. The relaxation mechanism is described using the principle of spring force. Finally, using various kinds of adaptive relaxation operators, the evaluation experiment for the stochastic model is presented, arriving at the relaxation operator, which is more useful than those in the past.