Abstract: A robust digital watermarking algorithm based on Zernike moments and DCT transformation is proposed, which estimates angles of rotation to promote the ability of anti-rotation attack. The orthogonal Zernike polynomials based on image content are made as watermarking signal, which is embedded into the middle or low frequence control by Watson’ perceptual model. Relevant detection method are used to identify the existence of watermarking. Experimental results show this algorithm has better performance of rubustness, nonvisibility, and testability.

Key words: digital watermarking, Zernike moment, rubustness

摘要： 提出一种基于Zernike矩和DCT变换的鲁棒数字水印算法，对图像的旋转角度进行估计，以提高其抵抗旋转攻击的能力，将基于图像内容的具有正交特性的Zernike多项式作为水印信号，并在Watson视觉感知模型的控制下嵌入图像低频系数中，同时采用相关检测方法确定水印的存在，实验结果表明，该算法具有良好的鲁棒性、不可见性及可检测性。

关键词: 数字水印, Zernike矩, 鲁棒性

CLC Number: 

• TP301.6