论文部分内容阅读
本文提出了一种基于奇异值分解和小波包分解相结合的全新水印算法。综合利用奇异值和方差的特征来对宿主图像进行预处理之后,提取出两个具有不同掩蔽效应的子图,分别对子图的奇异值和小波包系数使用抖动调制方法嵌入不同强度的水印。本文算法不仅实现了水印的盲提取,而且实现了水印鲁棒性与透明性的最佳折中。本文的另一创新之处是提出了一种基于神经网络的新的水印复原方法。实验结果表明本文方法较传统方法而言有更高的灵活性和鲁棒性,尤其具有很强的抗JPEG压缩能力。
This paper presents a new watermarking algorithm based on singular value decomposition and wavelet packet decomposition. After preprocessing the host image by using the features of singular value and variance, two subgraphs with different masking effects are extracted. The singular values and the wavelet packet coefficients of the subgraph are respectively jitter modulated to embed watermarks with different intensities. This algorithm not only implements blind extraction of watermark, but also realizes the best compromise between robustness and transparency of watermarking. Another innovation of this paper is to propose a new watermark recovery method based on neural network. The experimental results show that the proposed method is more flexible and robust than the traditional methods and especially has strong anti-JPEG compression ability.