【作者】 田云凯；

【导师】 贾传荧；

【作者基本信息】 大连海事大学 ， 交通信息工程及控制， 2007， 硕士

【摘要】 信息隐藏是近年来信息安全和多媒体信号处理领域中提出的一种解决媒体信息安全的新方法。它通过把秘密信息永久性地隐藏在可公开的媒体信息里，达到证实该媒体信息的所有权归属和数据完整性或传递秘密信息的目的，从而为数字信息的安全问题提供一种新的解决方法。 本文以灰度图像作为研究对象，首先介绍了信息隐藏技术的基本知识。它主要包括：信息隐藏技术的分类、信息隐藏的原理、信息隐藏的常见算法、信息隐藏技术与传统密码学的区别。 针对图像的信息隐藏问题，图像置乱是基础性的工作。本文第二部分先介绍了置乱变换的特点及其在图像信息隐藏中的作用，然后介绍了一种典型的置乱变换——Arnold变换，利用矩阵变换推导出了求Arnold反变换的公式，将Arnold变换运用到各种不同图像的置乱，最后用置乱程度验证了Arnold变换对于其它置乱变换的优势。 接着介绍了小波理论及其在图像水印中的应用。为后面的研究工作打下坚实的理论基础。 结合数据融合技术，本文给出了一种在小波域对两幅图像的低频部分进行分块的图像信息隐藏的算法。先对分快后的低频图像块进行置乱加密，然后利用各图像块能量的不同对不同的置乱图像块进行图像信息的隐藏，最后通过实验验证了隐藏图像的鲁棒性并通过实验数据的比较，指出了其对于其它算法的优势和不足。更多还原

【Abstract】 Information hiding is a new technique for information security and multimedia signal processing, which embeds the cryptical information into the open media information forever to testify the ownership of the media information and integrality of the data, consequently offers a new resolvable method for digital information security.Based on gray images, firstly we gives a brief introduction about the basic knowledge of the information hiding technique, which mainly includes the types, principle, common algorithms and difference on cryptography of information hiding technique.Since image scrambling is the basic work for image information hiding problem. Firstly this paper introduces the characteristic of scrambling transform and the function in image information hiding, and then the paper introduces a typical scrambling transform-Arnold transform. Through calculating with matrix transform, we derived the algorithm of anti-Arnold transform and apply it to all kinds of image scrambling.Finally we testify the superiority of Arnold transform to the other scrambling transform with scrambling degree.We introduce in succession the wavelet transform and the application in image watermark, which lay the stability foundation for the following study.According to the data fusion technique, the last part of this paper gives a method that the two images’ low frequency portions are divided into segments of the same size for the purpose of image information hiding. At first, the blocked low frequency segments are scrambled and encrypted, then the different scrambled segments are embedded with cryptical image information according to the difference of the image segments’ energy. In conclusion, the robustness of the algorithm is improved with numerical experiments, the advantage and the disadvantage are found as well through the comparison of the experiment data.更多还原