基于非方阵图像DWT-SVD的盲水印算法研究与仿真  

Research and Simulation of Blind Watermarking Algorithms Based on DWT-SVD of Non-Square Image

作　　者：张帅[1] 杨雪霞[2] ZHANG Shuai;YANG Xue-xia(Dept.of Teaching,Taiyuan Radio&TV University,Taiyuan 030024,China;School of Applied Science,Taiyuan University of Science and Technology,Taiyuan 030024,China)

机构地区：[1]太原广播电视大学教学研究中心,山西太原030024 [2]太原科技大学应用科学学院,山西太原030024

出　　处：《中北大学学报（自然科学版）》2020年第1期49-54,共6页Journal of North University of China(Natural Science Edition)

基　　金：国家自然科学基金资助项目（11602157)

摘　　要：为了解决水印算法仅对方阵图像研究的局限性,以及提取的水印信息在剪切攻击方面的不足,提出了一种在离散小波变换域内能够对任意尺寸图像进行变换,并结合矩阵奇异值分解的盲水印算法.通过对载体图像的规范化处理,使得图像可以进行任意级的小波分解,之后利用在奇异值中嵌入双重水印的思想解决了剪切攻击方面的问题.实验结果表明,该算法能够实现水印信息的盲提取,面对多种类型的攻击表现出较强的鲁棒性,并且在剪切攻击方面提取的水印信息比同类算法表现出更好的性能.In order to solve the limitation of watermarking algorithm,only the array image was studied and the shortage of extracted watermark in the cropping attack,a blind watermarking algorithms was proposed,which can transform any size image in the discrete wavelet transform domain and combine with matrix singular value decomposition.By normalizing the carrier image,the image can be decomposed by any level of wavelet transform,then the problem in shearing attack was solved by embedding double watermarking in singular value.The experimental results show that the algorithm can achieve blind extraction of watermark information,and it has strong robustness against many types of attacks,and the watermark information extracted in the shear attack performance is better than similar algorithms.

关 键 词：离散小波变换 数字水印 奇异值分解 鲁棒性 盲提取 

分 类 号：TN919.81[电子电信—通信与信息系统]