【作者】 张艳芬；

【导师】 孔俊；

【作者基本信息】 东北师范大学 ， 计算机软件与理论， 2008， 硕士

【摘要】 近年来,随着现代通信技术的飞速发展和计算机互联网技术的迅速普及,信息的共享、传播变得尤为重要。它给人们带来种种便利的同时也带来了一系列的信息安全问题,比如信息窃取问题、多媒体信息的版权保护问题、信息篡改问题等。而传统信息安全保障手段—密码术并不能真正解决此类问题。数字密写技术将秘密信息隐藏于公开的载体中进行传输,不易引起攻击者注意,降低了被截获的可能。现有的数字密写算法存在一定的缺陷,大多采用灰度图像作为传输的载体,限制了应用领域;采用一对一的嵌入,即一个秘密图像隐藏在一幅载体图像中,含密图像如果在传输过程中被截获或者被破坏,接收方将得不到任何信息。本文针对现有密写算法的不足之处,提出了一种新的基于小波变换的秘密图像共享方法。该算法弥补了经典的( r ,n)限方案的不足之处,通过设置多个门限值来控制恢复得到的秘密图像的质量。门限值设为( r1 ,r2,...,rk ,n),将秘密图像共享为n个子图,当拥有的子图数num≥rk时,恢复全部的秘密信息;当num < r1时,得不到任何秘密信息;当r1≤num<rk时,拥有的子图越多,恢复得到的秘密图像质量越好。采用基于块的Arnold变换置乱算法不仅保证了块内像素间的相关性,同时增强了秘密信息传输的隐蔽性和安全性;采用小波变换和位平面扫描技术保证了每个像素块越前面的像素越能体现原图的主要能量;采用最低有效位替换方法将各个子图嵌入到彩色图像R,G,B三个颜色通道中,达到了更好的视觉效果。实验结果表明本文提出的方法具有较好的不可见性、较高的安全性和较大的嵌入容量等特点。更多还原

【Abstract】 With the rapid development of modern communication technology and the wide application of computer internet technology, information sharing and transmission becomes more and more important. It has brought us great convenience and a great number of information security problems as well. Such as information theft, copyright protection of multimedia data and image authentication. But traditional security protection way-cryptography can’t indeed solve the above problems. Steganography hides the secret information into public carrier to transmit it secretly, which can not arouse attacker’s attention and reduce the possibility of being intercepted and captured.There is some fault in recent Steganography, in which gray-image is used as carrier, which restricts the application domain. Further more, one secret image is hidden in one carrier image, the receiver couldn’t obtain anything once the stego-image is attacked or destroyed.This paper brings forward a new secret image sharing method based on discrete wavelet transform according to the fault of recent methods. This method is reparation of traditional image sharing technique, which controls the vision quality of the extracted secret image by defining several thresholds. The thresholds are defined as ( r1 ,r2,...,rk ,n). Sharing the secret image into n shadows, when the number of shadows are more than rk , the receiver can obtain all the information of the secret image, when fewer than r1 , we can get nothing, when the number between the two values, the more shadows, the better quality of the secret image.Adapting Arnold transform based on block not only ensure the relativity between pixels, but also improve the invisibility and the security of the secret information. discrete wavelet transform and Bit-plane scanning is used to ensure the more information the frontier pixels of each block pixels. A well visual effect can be achieved adapting Least Significant Bit substitution method to imbed each shadow image into R, G and B channels of the cover image. The experimental results show that our method can achieve the better invisibility, higher security and the big imbedding space.更多还原