本文提出一种新的基于离散小波变换（DWT）、离散余弦变换(DCT)和矩阵奇异值分解(SVD)的鲁棒水印算法。首先按照本文提出的小波分解准则对载体图像进行四层小波分解，取第四层的低频子图与三个高频子图；同样对水印图像进行小波分解得到低频子图与三个高频子图。然后用DCT、SVD方法,结合本文提出的相互嵌入准则将水印图像的低频子图和三个高频子图自适应地嵌入到载体图像的低频子图和三个高频子图中。最后作SVD、DCT和DWT逆变换得到含水印图像。在该方法中，置乱技术用于水印图像隐藏的预处理和后处理，更加保障了数据的安全性。实验结果表明，本算法具有较强的鲁棒性。

A novel watermarking algorithm for digital images based on DWT,DCT and SVD is proposed in this paper.First, according to the decomposition criteria proposed in this paper,we apply a four layer DWT on the original image,and choose the lowfrequency subimage and the three highfrequency subimages of the fourth layer.Similarly,we apply DWT to the watermaking image and get the lowfrequency subimage and the three highfrequency subimages.Then,by using DCT,SVD and the decomposition criteria proposed in the paper,we embed the lowfrequency subimage and the three highfrequency subimages obtained from the watermarking image into those of the original image adaptively.Finally the watermarked image is obtained after the inverse SVD,DCT and DWT.In the scheme,scrambling technology is used for preprocessing and postprocessing the watermarking image's information hiding,and it also well guarantees the safety of data.Experimental results demonstrate that this method is robust.