Novel blind non-additive robust watermarking using 1-D chaotic map

We propose a novel blind non-additive robust watermarking scheme in this paper. The energy of the watermark is spread to all regions of the host data instead of some individual elements, which entitles the watermark with imperceptibility and high robustness. A class of 1-D Markov chaotic maps employed to perform host data element classification and watermark encryption ensures the security of the system. To prove the validity of this proposed scheme, some analyses and brief objective comparisons with the popular spread spectrum (SS) scheme are also presented. Simulation results show that our scheme can survive severe processing such as high-ratio JPEG compression, Gaussian noise pollution and histogram equalization.