A Watermarking Method Based on Optimization Statistics

Abstract

This paper presents a robust image watermarking scheme based on optimization Statistics. This method aims to select a feature space, which has the greatest robustness against various attacks after watermarked. We separately obtain the feature spaces through calculating the statistical property of the digital image. Passing spaces decomposing and reconstructing of the feature spaces, constructing the embedding matrix, we obtain that the robustness of the approach lies in hiding a watermark in the subspace that is the least susceptible to potential modification; and realize the optimization statistics of the embedding watermark. Through analysis and constraint the conditions of subspace, the algorithm we proposed can obtain a high detection probability and security, a low false alarm probability. Experimental results show that the proposed scheme is robust against a kind of attacks.

DOI: http://dx.doi.org/10.11591/telkomnika.v11i8.3142