摘要
针对多媒体图像版权保护,提升抵抗强剪切攻击能力,提出一种基于非负矩阵分解NMF(Non-negative Factorization)和改进的Sobel算子特征提取的零水印方案。方案以抗强剪切攻击为核心,采用Fibonacci变换和混沌序列置乱技术对水印进行加密预处理;对原始图像进行NMF分解,得到系数矩阵并注册在版权中心;采用改进Sobel算子边缘提取算法构造特征向量,并与水印异或得到原始图像的版权信息。实验结果表明,当剪切面积达到96.50%以上,方案都能有效抵抗强剪切攻击,同时对缩放、加噪、滤波、JPEG压缩等攻击具有良好的鲁棒性。
For copyright protection of multimedia resources and the ability enhancement of zero watermark algorithm to resist strong shearing attack,this paper proposes a zero watermark scheme based on NMF and improved Sobel Operator edge detection.In this scheme,the anti-strong shear attack is the core.Fibonacci transform and chaotic sequence scrambling technology are used to pre⁃process the watermark.The original image is decomposed with NMF,and the coefficient matrix is registered in the copyright cen⁃ter.The improved Sobel Operatoredge detection is used to construct the feature vector;which is then operated with watermark by XOR operation to obtain the copyright information from the original image.The experimental results show that when the shearing area reaches to 96.50%,the scheme can extract the watermark information well,and also has good robustness against scaling,noise,filtering,JPEG compression,and other attacks.
作者
陈亚男
殷柯欣
朱建启
包芳
CHEN Ya-nan;YIN Ke-xin;ZHU Jian-qi;BAO Fang(School of Computer Science and Engineering,Changchun University of Technology,Changchun 130012,China;School of Com-puter Science and Technology,Jilin University,Changchun 130012,China)
出处
《电脑知识与技术》
2020年第25期15-18,共4页
Computer Knowledge and Technology
基金
国家自然科学基金项目“车载自组网服务质量与安全性能跨层联合优化理论与技术研究”(61572229)。