基于DCT的自适应多重彩色图像盲水印算法  

Adaptive Color Image Blind Multi-watermark Algorithm Based on DCT

作　　者：谢斌[1] 刘珊[1] 任克强[1] 

机构地区：[1]江西理工大学信息工程学院,江西赣州341000

出　　处：《电视技术》2014年第9期21-24,共4页Video Engineering

基　　金：江西省教育厅青年科学基金项目(GJJ11132)

摘　　要：针对现有图像水印算法的一些不足,提出了一种基于离散余弦变换的自适应多重彩色图像盲水印算法。该算法根据彩色载体图像内容特点,将预处理后的不同二值图像水印自适应地嵌入到其绿色分量的离散余弦变换低频系数和中频系数,实现了不同水印在彩色载体图像中的嵌入及盲提取。实验结果表明,该算法在保证不可觉察性的前提下能够较好地抵抗各类常规攻击,具有良好的稳健性。An adaptive multi-color image blind watermarking algorithm is proposed based on discrete cosine transform for some shortcomings of the existing image watermarking algorithms. According to the color vector image content features,after pretreatment of the type of binary image watermark is adaptively embedded into the green component of the discrete cosine transform low-frequency coefficient and intermediate frequency coefficients. the type of watermark embedding and blind extraction of the color vector image are implemented. The experimental results show that the algorithm under the promise of not aware can resist all kinds of conventional attack better,and it has a good robustness.

关 键 词：ARNOLD变换 多重图像水印 离散余弦变换 自适应 盲提取 

分 类 号：TN911.73[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]