Error analysis for an RST invariant image watermarking algorithm

We have proposed a rotation, scaling and translation (RST) invariant image watermarking algorithm, where the cover image is mathematically approximated using the mixture Generalized Gaussian distribution. In this paper, we propose an error analyzing model for the RST invariant watermarking algorithm to guide the determination of the watermarking parameters. The stochastic analysis for the probability error including the false positive probability and the false negative probability of the RST invariant image watermarking algorithm is elaborated in terms of the relative entropy and the PSNR of the watermarked image. The experimental results demonstrated the effectiveness of the proposed model.