首页 | 本学科首页   官方微博 | 高级检索  
     

数字水印嵌入强度最优化分析
引用本文:王树梅,赵卫东,王志成.数字水印嵌入强度最优化分析[J].计算机安全,2007(10):1-4.
作者姓名:王树梅  赵卫东  王志成
作者单位:同济大学,CAD研究中心,上海,201804;徐州师范大学,计算机学院,江苏,徐州,221116;同济大学,CAD研究中心,上海,201804
基金项目:国家高技术研究发展计划(863计划) , 上海市科委资助项目
摘    要:数字水印技术是一种保护版权的重要技术.水印系统里,最重要的两个关键因素就是水印嵌入后,载体图像的抗攻击性(鲁棒性)和不可见性.如何在水印嵌入过程中,使得这样一对矛盾达到统一,嵌入强度α起着较为关键的作用.α过大会影响水印的不可见性;α过小,水印的抗攻击性减弱,从而失去嵌入水印的本意.该文提出一种基于载体图像和水印信息本身特点的求解α的算法,该算法详细分析了图像特点与鲁棒性、不可见性之间的关系.通过分析图像和含水印图像的相似度以及峰值信噪比,计算出α的解.实验证明,通过此方法求得的α能够更好地适应原始图像,使得水印图像的鲁棒性和不可见性达到最佳折中.

关 键 词:数字水印  嵌入强度  最优化  离散余弦变换
修稿时间:2007-06-11

An Optimization Algorithm of Digital Watermarking Scaling Parameter
WANG Shu-mei,ZHAO Wei-dong,WANG Zhi-Cheng.An Optimization Algorithm of Digital Watermarking Scaling Parameter[J].Network & Computer Security,2007(10):1-4.
Authors:WANG Shu-mei  ZHAO Wei-dong  WANG Zhi-Cheng
Affiliation:1. CAD Research Center, Tongji University, Shanghai 201804 P.R. Chine, 2. Computer Academy, Xuzhou Normal University, Xuzhou 221116 P, R. China
Abstract:Digital watermarking is an important technology of copyright protection. In watermarking system, two key factors are the attack-resistant (robustness) and invisibility of the watermark embedded image. How to make the contradiction to reunification in the embedding process, plays a crucial role. it is too great to get visiblity,and too small, the robustness is weakened, and thus losing embedded watermark intention. In this paper, an algorithm to get best scaling parameter is presented based on the characteristics of the carrier image, in which the relationship between characteristics and robustness of carrier is analyzed detailed. The scaling parameter is found through analyzing similarity and peak signal to ratio between the original image and the watermark image. Many experiments are evidences that this method can make scaling parameter to better adapt to the original image,and also make robustness and invisibility to achieve the best compromise.
Keywords:digital watermarking  scaling parameter  optimization  discrete cosine transform
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号