Image Watermarking Resistant To Combined Geometric And Removal Attacks

In this paper, we propose an improved autocorrelation function (ACF)-based water21 marking that is robust against combined geometric and removal attacks. ACF-based watermarking is known to be effective against geometric attacks. In this watermark23 ing scheme, both the embedded watermark and autocorrelation peaks of the watermark should survive a variety of attacks. However, the autocorrelation peaks are not strong 25 enough. The proposed method improves the robustness of ACF-based watermarking by enhancing the strength of the autocorrelation peaks. Instead of the simple addition of a 27 watermark to an image, the proposed watermark embedder uses the information of an original image more actively during the embedding process, so that the marked image 29 has strong autocorrelation peaks. Experimental results show that the proposed scheme provides better robustness than conventional ACF-based watermarking against a variety 31 of combined geometric-removal attacks.