Abstract: This paper proposes a new algorithm based on Nonsampling Contourlet Transform（NSCT） and Discrete Cosine Transformation（DCT） to improve the watermarking’s anti-rotation attack robustness and the embedding quantity. It transforms the image with NSCT, then does the low-frequency coefficients with DCT, decomposes the low-frequency coefficients of DCT domain with Singular Value Decomposition（SVD）. Experimental results show that the algorithm has a good robustness in filter and rotation, and extracts the watermarking very well in rotation attack especially.

Key words: digital watermarking, singular value decomposition, discrete cosine transformation, nonsampling Contourlet transform

摘要： 为了提高水印的抗旋转攻击鲁棒性，加大水印的嵌入量，提出了一种基于非采样Contourlet变换（NSCT）和离散余弦变换（DCT）结合的双重变换域水印算法。对图像进行NSCT变换，将低频系数进行DCT变换，对DCT域低频系数进行奇异值分解（SVD）。经仿真实验表明，该算法对常规滤波、旋转有很好的鲁棒性，特别面对旋转攻击时，算法仍能很好地提取出水印信息。

关键词: 数字水印, 奇异值分解, 离散余弦变换, 非采样Contourlet变换