【作者】 周建伟；

【导师】 张素文；

【作者基本信息】 武汉理工大学 ， 控制理论与控制工程， 2007， 硕士

【摘要】 随着计算机网络和多媒体技术的飞速发展，多媒体数据逐渐成为人们获取信息的重要来源，并成为人们生活的重要组成部分。然而当前数字信号处理和计算机网络传输技术的迅猛发展却使数据的版权保护面临严峻的挑战。因而，如何在开放的网路环境下对数字作品实施有效的版权保护已成为国际上研究的热门话题。数字水印技术是当前信息隐藏技术研究最为活跃的一个分支。数字水印技术是依据某种的算法将特定的数字信息添加到载体文件中，并且通过一定的手段和检测标准，提取或者检测它们。本文讨论的是应用信息隐藏技术的理论，在声音文件中添加数字水印的技术。首先，本文介绍了数字水印问题的提出及其相关概念，阐述了本课题的研究背景和意义。其次，阐述了音频数字水印算法的基本原理、模型及实现过程。然后，在分析前人的工作和当前水印发展的状况的基础上，本论文通过分析数字音频水印和人类听觉系统的特点，提出了一种基于小波变换的数字音频水印算法。这种音频水印算法是一种基于离散小波变换(DWT)的的音频水印算法。该算法首先对待嵌入水印的数字音频信号进行分帧、分节预处理，再选择每帧的前两节进行离散小波变换，根据人类听觉系统的时域掩蔽效应进行能量对比找到嵌入点，进行数据的嵌入。该算法嵌入水印时并不直接将图像嵌入，而是先对图像进行降维预处理，把预处理后图像序列嵌入到音频信号中。通过实验对其性能进行了主观评价和客观数据分析，考察了水印的透明性攻击的鲁棒性，实验证明，上述算法具有较强的健壮性，较好的透明性，提取水印属盲水印提取，能经受添加噪声、重采样、低通滤波、重新量化、音频格式转换等常见信号处理及攻击。更多还原

【Abstract】 The digital media has become a main way for information communication along with the rapid development of digital technology and computer networks. However, the currently rapid development of digital signal processing and Internet transmission techniques have threatened the copyright of digital data severely. Therefore how to effectively protect the copyright of digital products in an open network environment becomes an attractive topic. Digital watermark is the most active of the hiding technologies. The technology embeds some data into a carrier data, and this data can be checked with some methods.This article discusses is the application information hideaway technology theory, increases the digital watermark in the sound document the technology. First, this article introduced digital watermark question statement and the correlation concept, elaborated this topic research background and the significance. Next, elaborated the audio frequency numeral watermark algorithm basic principle, the model and the realization process. Then, in analyzes predecessor’s work and in the current watermark development condition foundation, the present paper through the analysis numeral audio frequency watermark and the human sense of hearing system characteristic, proposed one kind based on the wavelet transformation digital audio frequency watermark algorithm.This kind of audio frequency watermark algorithm is one kind transforms (DWT) based on the separate wavelet the audio frequency watermark algorithm. This algorithm first treats the inserting watermark the numeral tonic train signaling to carry on a minute frame, the sectionalization pretreatment, again chooses each frame first two to carry on the separate wavelet transformation, carries on the energy contrast according to the human sense of hearing system time domain masking effect to find the inserting spot, carries on the data inserting. When this algorithm inserting watermark not directly inserts the image, but is carries on first to the image falls Uygur to pretreat, after the image sequence inserts the pretreatment to the tonic train signaling in, Has carried on the subjective appraisal and the objective data analysis through the experiment to its performance, inspected the watermark transparent attack robustness, the experiment had proven, the above algorithm had the strong toughness, the good transparency, the extraction watermark was the blind watermark extraction, could undergo the increase noise, the heavy sampling, the low pass filter, again the quantification, the audio frequency format conversion and so on the common signal processing and the attack.更多还原