Reversible Data Hiding Based On Histogram Modification And Watermark Scheme Based On SVD

Reversible data hiding based on histogram modification is widely studied and has a wide range of applications because of its low computation complexity, high capacity and reversible property. The key point of the error prediction based algorithm is to design accurate prediction methods, more accuracy prediction leads to higher capacity. In the problem of copyright protection, digital watermark based on singular value decomposition (SVD) embeds watermark through modifying singular vectors or singular values. When the watermarked image is damaged in some extent the singular value of SVD may keep stable.Binary tree structure solves the problem of communicating pairs of peak points. The triangle predictor and error energy estimator improve the image quality and payload. Due to the inaccuracy of the triangle predictors, in this paper, a predictor which uses neighbourhood information is proposed. The error energy and the way to extract the secret message are accordingly modified. The experimental results reveal that the proposed method has a more precise predictor and an higher capacity and it’s more confidential.In the experiment of the 7 most common pictures the capacity has improved by an average of 8.7%. In the same situation, the capacity has improved by 300%.About the water mark algorithm, in this paper, we first analyse the algorithms based on SVM and design a process which may get a watermark in a picture without watermarks. That means, there is some problem in the SVM based watermark algorithms which leads to the unreliability. In this paper, we proposed a scheme based on SVD. In this method, an image is partitioned to blocks, and then some blocks are chosen to embed watermark according to the information of edges. The transparency and robustness are considered in evaluating the performance of watermark algorithms. And the PSO algorithm is used to get the parameter adaptively for an optimized effect. Analysis and experiments reveal that the scheme proposed is reliable, has a low computation complexity, and the transparency and robustness are also high.