OPTIMAL LOCATION OF RANDOM TARGETS IN RANDOM BACKGROUND - RANDOM MARKOV-FIELDS MODELIZATION

The problem of the optimal location of a target with random texture ap pearing on a random background is addressed. We analyze the case where the texture of the target and of the background can be described by t he same Markov random field model. We show that the optimal processor consists in preprocessing the input image and then performing two corr elations. This result opens new perspectives for optimal optical corre lation.