基于DCT和CNN混沌系统的彩色数字水印加密新算法  

New Algorithm of Color Digital Watermark Encryption Based on DCT and CNN Chaotic System

作　　者：赵国敏[1] 李国东[1] 朱文辉[2] 

机构地区：[1]新疆财经大学应用数学学院,乌鲁木齐830012 [2]新疆财经大学经济学院,乌鲁木齐830012

出　　处：《四川理工学院学报（自然科学版）》2014年第5期58-63,共6页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

基　　金：国家教育部人文社会科学基金(13YJAZH040);新疆维吾尔自治区高校科研计划项目(XJEDU2013I26)

摘　　要：数字水印技术作为抵抗多媒体盗版的最后一道技术防线,具有广泛的应用前景和巨大的经济价值。基于离散余弦变换(DCT)以及细胞神经网络(CNN)混沌理论提出了一种数字水印加密新算法。算法分两步进行,首先是利用5阶细胞神经网络混沌系统产生的随机序列辅助某种运算对彩色水印图像加密,然后利用分块离散余弦变换将加密以后的彩色水印图像嵌入到载体彩色图像中,以此来实现水印加密以及嵌入的过程。在仿真实验基础上,通过指标PSNR和NC的定量分析,结果证明新算法具有较强鲁棒性,不可感知性和安全性。As the last line of defense technology against the multimedia piracy,digital watermarking technology has wide application prospect and great economic value.A new digital watermarking encryption algorithm is proposed based on the discrete cosine transform（ DCT） and cellular neural network（ CNN） chaotic theory.The algorithm is consisted of two steps to complete the process of watermark encryption and embedding.Firstly,the random sequences are generated by the 5 order cellular neural network chaotic system and are operated to achieve the encryption of color watermark image.Then the encrypted color watermark image is embedded into the host image by the block method of discrete cosine transform.At last,the simulation experiments are done,and the results show that the watermarking algorithm has strong robustness,imperceptibility and security through the quantitative analysis of the indexes of PSNR and NC.

关 键 词：数字水印 离散余弦变换(DCT) 细胞神经网络(CNN) 超混沌 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]