矢量量化压缩图像中的安全隐写方法  

Secure steganographic method for vector quantization-compressed images

作　　者：刘光杰[1] 戴跃伟[1] 王执铨[1] 杨静宇[2] 

机构地区：[1]南京理工大学自动化学院,南京210094 [2]南京理工大学计算机学院,南京210094

出　　处：《控制与决策》2008年第11期1243-1248,共6页Control and Decision

基　　金：国家自然科学基金项目(60374066);中国博士后基金项目(20070421017)

摘　　要：提出一种用于矢量量化压缩图像的安全数据隐藏方案.为降低数据嵌入引入的失真,以码字间的矢量均方差为优化指标,采用遗传算法实现码本的优化分割,并提出基于码本分割的数据嵌入算法.采用基于自适应算术熵解码的数据映射方法,实现了嵌入前后统计特性的保持.实验结果表明,所提出的算法在容量、失真水平和安全性方面具有较好的综合性能.A secure data hiding scheme for vector quantization （VQ）-compressed images is proposed in this paper. To decrease the distortion caused by data embedding, the vector mean square error between codewords is taken as optimization index, the genetic algorithm is used to realize the partition of codebook, and the data embedding approach is given based on codebook partition. Meanwhile, the adaptive-arithmetic-decoder-based data mapping method is used to keep the statistical characteristics before and after data hiding. The experimental results show that the proposed scheme has better combined performance in capacity, distortion level and security.

关 键 词：数据隐藏 矢量量化 遗传算法 相对熵 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]