基于DWT与SVD自适应公开水印技术

A Image-Adaptive Public Watermarking Technique based on DWT-SVD

在深入研究小波变换理论、奇异值分解理论的基础上,通过计算表达式来建立水印嵌入参数(如嵌入强度等)和图像质量的关系,提出了一种基于DWT与SVD鲁棒自适应的水印算法,算法中选取小波域的LL子带的最大奇异值自适应量化的方法嵌入水印,并能够盲提取,实验表明,本算法对JPEG压缩、中值滤波、亚抽样、低通滤波等图像处理有很强的稳健性。

Based on block DWT Decomposition and Singular Value Decomposition, we proposed a new algorithm with the method of embedding the watermarking into the largest singular value. Meanwhile we produced a formula to estimate the PSNR. Experiments and attack test show that the algorithm has good robustness on JPEG, median filter, sub-sampling, Gaussian low pass filter etc.