Optimization of Natural Watermarking Using Transportation Theory

摘要

This article presents a new approach to use secure spread-spectrum watermarking in the Watermarked Only Attack (WOA) framework (the adversary owns several marked contents and wants to estimate the secret key). Because security of watermarking schemes relies on the distribution of marked contents in a private subspace, we use the transportation theory in order to derive a mapping that will minimize the global distortion of Natural Watermarking (NW) while keeping the same security level. Contrarily to another technique presented in [13], the embedding does not depend on computationally demanding pre-computed mapping. Moreover, we show that this method does not reduce the robustness of the classical NW scheme and enables to decrease the WCR by several dBs regarding former implementations. Tests are made on 2000 Gaussian signals.