摘要
提出了一种可自恢复水印的半盲数字水印技术,采用了双重数字水印的方法,建立了一套独有的自恢复系统来获取数字水印技术所必须具备的高稳健性。人类视觉模型在算法中的引入则为水印的透明性提供了有力的保障,该算法的抽取和检测不需要原始水印参与,是一种半盲水印算法。实验证实了该算法不仅可以有效地应对各种常规的信号处理攻击,而且还可以防止大部分恶意水印攻击,相比于其他多种同类型水印算法,其稳健性和透明性都有明显提高。
A kind of self - recovery and half - blind watermarking technology was proposed By adopting the method of dual water- marks, a unique self - recovery system to obtain great robustness needed in watermarking technology was built. The introduction of the human visual system offers forceful guarantee for the invisibility of watermarking. The extraction and detection of the algorithm is a kind of half - blind watermarking. There is no need for them to get involved in the original watermarking. The experiment results prove that the algorithm can not only effectively resist various regular attacks to signal processing attacks, but also can prevent major intentional watermarking attacks. Compared with other similar watermarking algorithms, the robustness and transparency of the proposed algorithm has been greatly improved.
出处
《武汉理工大学学报(信息与管理工程版)》
CAS
2008年第6期881-885,919,共6页
Journal of Wuhan University of Technology:Information & Management Engineering
基金
湖北省自然科学基金资助项目(2005ABA282)