DCT域自适应公开水印技术

Image-adaptive Public Watermarking Based on DCT

提出了一种新的DCT域自适应公开水印算法。与以往DCT域算法有所不同，该文算法通过调整两个不同的DCT分块内相对应中频系数间大小关系以嵌入水印。算法首先将原图像各8×8DCT块按Hilbert扫描顺序排列，然后每次在原图像Hilbert块索引序列中伪随机选取两个相邻8×8块，在充分考虑人类视觉系统（HVS）特性的基础上，结合块视觉因子，通过不同强度调整两相邻DCT块中对应中频系数间的大小关系，以嵌入水印信号，嵌入水印与原图像是相自适应的。实验结果表明，该算法是在保证水印不可见的同时，对常见的信号处理操作和一些几何扭曲是鲁棒的，特别是对锐化、对比度增强、边缘增强和直方图均衡等信号增强处理，其鲁棒性更为可观。

A new image-adaptive public watermarking based on DCT is presented. Being different from the most previous DCT-based algorithms, the proposed technique embeds a watermark by modifying the polarity of the corresponding coefficients in two different DCT blocks other than changing the coefficients in the same block. First, the 8×8 DCT blocks of the host image are rearranged into a 1-D Hilbert sequence in Hilbert scanning order, then two neighboring blocks in Hilbert sequence is pseudo-randomly selected by using chaotic sequences. A watermark is embedded into the original image by changing the polarity of the corresponding middle-frequency coefficients in the two chosen neighboring DCT blocks, and the watermark is adapted to the host image by exploiting the masking characteristics of the human visual system to ensure the watermark invisibility. The experimental results show that the algorithm is robust to common signal processing techniques and some geometric distortions. Especially, it achieves high robustness under signal enhancement operations, such as sharpening, contrast enhancement, edge enhancement, histogram equalization, and so on.