摘要： 利用树状小波分解结合人类视觉系统的特性，提出一种基于数据融合的鲁棒性数字水印算法，可以向载体图像中自适应地嵌入多个数字水印副本。在每个块中使用独立分量分析的方法提取水印，对提取出的多个水印副本图像进行融合操作，以提高水印的鲁棒性。仿真试验表明，以该方法嵌入的水印具有较高的透明性，对常见图像处理攻击有很强的鲁棒性。

关键词: 数字水印, 树状小波分解, 独立分量分析, 数据融合

Abstract: By using data fusion, this paper proposes a robust digital watermarking algorithm. Multiple copies of digital watermarking are embedded into the carrier image by exploiting the characteristics of tree-structured wavelet decomposition and human visual system, and Independent Component Analysis(ICA) method can be used to extract these copies. These extracted copies are fused by PCA image fusion algorithm so as to improve the robustness of the algorithm. Simulation experiments show that the scheme is effective, and achieves higher transparency and strong robustness to common image processing attacks.

Key words: digital watermarking, tree-structured wavelet decomposition, Independent Component Analysis(ICA), data fusion

中图分类号: 

• TP301.6