块奇异值分解和量化实现的图像数字水印算法

讨论了当前将水印嵌入在块奇异值分解（SVD，singular value decomposition）后最大奇异值中的图像数字水印算法不足，进而提出了两种新的将水印嵌入在块SVD后最大奇异值除外的其余奇异值中的图像数字水印算法。两种新算法均采用了量化嵌入策略，从而使两种算法在提取水印时无需任何原始信息的帮助。实验结果表...

Blocked singular value decomposition and quantization based image digital watermarking algorithm

The shortage of the existing image digital watermarking algorithms that imbed watermarks into the largest singular values of sub blocks of the original image is analyzed and discussed.Then two novel image digital watermarking algorithms are proposed.The two algorithms imbed watermarks into the singular values of sub blocks of original image excluding the largest ones.Using the quantization scheme,the two new algorithms can extract watermarks without original information.Experimental results show that the proposed two new algorithms have good transparency of the embedded watermark image and are robust against attacks,such as brightness adjustment,contrast adjustment,slight rotation,slight translation and noise adding.