一种基于CT-SVD的鲁棒水印算法

提出了基于Contourlet变换(CT)和奇异值分解(SVD)的水印算法.与通常直接将水印嵌入至系数中的CT算法不同,选择了CT域低通子带的部分最大系数构建矩阵,对该矩阵进行奇异值分解,经Arnold置乱后的水印结合人眼视觉特性(HVS)被自适应地嵌入至奇异值中.实验结果表明,算法保证了极高的峰值信噪比(PSNR),并且体现了对缩放、JPEG压缩、剪切、滤波等攻击很好的鲁棒性.

A Robust CT-SVD Based Watermarking Scheme

In this paper, we present a watermarking scheme based on contourlet transform (CT) and singular value decomposition (SVD). Unlike usual CT algorithm where the watermarking is embedded directly into the coefficients, we select several maximum coefficients in the lowpass subband of CT to construct a matrix. Furthermore, the matrix is decomposed using SVD, and the watermarking image after Arnold scrambling is embedded into the singular values combined with human visual system (HVS) adaptively. Experimental results show that the proposed scheme guarantees high peak signal to noise ratio (PSNR), and is very robust against scaling, JPEG compression, cropping, filtering etc.