Research On Robust Reversible Watermarking Algorithm For Digital Image

In recent years,robust reversible watermarking have attracted many attentions.Robust reversible watermarking can correctly extract watermark information from the carrier images if only a certain degree of attack are confronted,and the carriers can be perfectly restored when they have not been attacked or edited after watermarking embedding.Robust reversible watermarking has been widely applied in military,judicial and medical fields for its reversibility and robustness.After carefully analyzing the existing robust reversible watermarking works,we propose two enhanced robust reversible watermarking algorithms and the main work of this thesis are as follows:Many space domain based algorithms split the carrier into many blocks and embed data into each block.These algorithms can achieve strong robustness at the expense of capacity.To solve this problem,we propose a re-embedding based algorithm which can improve the capacity of watermarking by taking both the robustness and embedding capacity into account.The proposed algorithm can improve the embedding capacity by shifting two independent statistics to embed information.The experimental results show that the proposed algorithm can increase the capacity of watermarking without damaging the robustness.We tackle the existing problems to improve the capacity and robustness of the wavelet based watermarking algorithms,and propose an Integer wavelet Transform and voting principle based watermarking algorithm,which can ensure that the information can be correctly exacted even more than half of the sub-watermarks have be attacked.The simulation results demonstrate that the robustness and capacity of the proposed algorithm are very promising.