Optimum decoding of non-additive full frame DFT watermarks

Abstract

The problem of optimum watermark recovery in a non additive, non Gaussian framework is addressed. Watermark casting is carried out on the frequency domain according to an additive-multiplicative rule. The structure of the optimum decoder is derived based on statistical decision theory. The Neyman-Pearson criterion is used to minimize the probability of missing the watermark for a given false detection rate. Experimental results highlights the superiority of the novel detector scheme with respect to conventional correlation-based decoding.