一种利用感知模型增加容量的新颖盲水印算法

感知模型在水印算法中有重要应用,它很好的解决了水印不可感知性和鲁棒性间的矛盾。但感知模型和水印容量间的关系还没有受到应有重视。本文提出一种新颖盲水印算法,利用感知模型增加水印容量,根据DCT系数噪声门限大小决定嵌入比特数,噪声门限越大,嵌入比特数越多。本算法可灵活地调节水印不可感知性、鲁棒性和水印容量间的关系。试验结果表明,算法是有效的。

A Novel Blind Watermarking Scheme Using Perceptual Model to Improve Capacity

Perceptual model has important application in watermarking schemes. It can solve the conflict between imperception and robustness well. But the relationship between perceptual model and watermarking capacity has not been considered enough. In this paper, a novel blind watermarking scheme has been proposed. The watermarking capacity has been obviously increased using perceptual model. The embedded bit number can be decided by calculating the noise threshold of DCT coefficients. The larger the threshold is, the more the embedded bits can be. The relationship among imperception, robustness and capacity can easily be adjusted by using the scheme. The experiment results show that the scheme is rather effective.