一种新的自适应数字水印算法

摘要

为了提高数字水印抗击各种图像攻击的性能和保持图像的稳健性和不可见性．提出了一种基于离散小波变换（DWT），SVD（singularvaluedecomposition）奇异值分解水印图像和原始载体图像的离散余弦变换（DCT）的自适应水印嵌入算法，主要是将水印图像的两次小波变换后的低频分量潜入到原始图像分块经过SVD分解的S分量矩阵中，同时根据图像的JPEG压缩比的不同计算各个图像块的水印调节因子。实验证明该算法在抗击JPEG压缩、中值滤波、加噪等均具有很好的鲁棒性，嵌入后的图像的PSNR达到38．具有良好的视觉掩蔽性%In order to improve the performance of digital watermarking to fight a variety of image attacks and to maintain the robustness of the image and visibility was proposed based on discrete wavelet transform (DWT), singular value decomposition SVD (singular value decomposition) of the watermarked image and the original carrier image discrete cosine transform (DCT) adaptive watermark embedding algorithm is mainly low frequency components in the two wavelet transform of the watermarked image dive into the original image block S-component matrix after SVD decomposition, according to the JPEG compression ratio different calculation for each image block watermark adjustment its factor. The experiments prove that the algorithm in the fight against JPEG compression, median filtering, noise will have great good robustness, the PSNR of the embedded image of 38 ,with good visual masking.