基于压缩传感的自同步图像水印算法

针对目前基于压缩传感水印算法存在的抗几何攻击性能不足的现状,提出了一种基于压缩传感的自同步图像水印算法.算法将水印信息和同步信号嵌入原始图像稀疏化后的观测域中,使水印具有自同步能力.同时利用观测值局部特性,有效提高了在观测域中搜索同步信号的效率,较好地解决了同步信号鲁棒性和搜索量之间的矛盾.实验结果表明,该算法对于常规图像处理攻击和几何攻击均具有较强的鲁棒性,同时抗提取实验表明,该算法能有效抵抗非法提取水印信息,增加了安全性.

Image Watermarking with Self-Synchronization Based on Compressive Sensing

In order to solve the shortage of present compressive sensing image watermarking resistant to geometric attacks, a novel image watermarking algorithm with self-synchronization based on compressive sensing is presented in this paper. The watermarking signal and the synchronization code are embedded into the measurement domain of image sparse coefficients, and then the watermarking has the ability of self-synchronization. By exploiting the localization property of measurement value to decrease the computational load for searching synchronization codes effectively, we resolve the conflicts between robustness and computational load for searching synchronization codes. Experimental results show the proposed algorithm is robust to common image processing and geometric attacks; more importantly, the anti-extraction experiments show that it can efficiently resist the illegal extraction of the watermark, and thus the security is guaranteed.