GENERALIZED OPTIMUM RECEIVER FOR PATTERN-RECOGNITION WITH MULTIPLICATIVE, ADDITIVE, AND NONOVERLAPPING BACKGROUND-NOISE

We design an optimum receiver to detect a pattern or a reference signa l. We design a receiver that detects the signal distorted by a multipl icative noise on the signal itself, as well as by additive noise and b y nonoverlapping scene noise. We design the optimum receiver under the condition in which the statistics of the multiplicative and nonoverla pping scene noise are not available. In the case in which additive noi se is present and the statistics of the multiplicative noise are not k nown, the usual method of maximizing the likelihood function to estima te the statistics of stationary noise fails. We overcome this problem by viewing the noise processes as vector random variables and describe two different schemes to estimate the statistics of the multiplicativ e noise. Using computer simulations we show that, for the images teste d here, the optimum receiver performs better than some of the existing receivers. (C) 1998 Optical Society of America.