Abstract
Detector based on maximum a posteriori probability (MAP) criterion has been introduced for identifying image watermark in the discrete cosine transform domain. This type of detector has been shown to be optimum and robust to common image processing operations. In this paper, an MAP detector in the discrete wavelet transform domain is proposed. It is based on modeling the wavelet coefficients by a generalized Gaussian distribution. Numerical experiments are used to compare the effectiveness of the proposed detector with the conventional correlation detector.