基于数字水印的可恢复语音取证技术研究

发布时间：2018-06-28 06:03

本文选题：数字水印 + 语音取证　；参考：《信阳师范学院》2017年硕士论文

【摘要】：数字信号处理技术的发展和高速网络的普及,使数字信号成为了我们获取和交换信息的主要载体,例如被二值化的数字语音信号。数字语音信号具有易存储、易复制和易传输等优点,作为信息载体广泛应用在电话通讯、新闻报道等方面。同时还可作为电子证据,应用在医疗记录和法庭举证等环节。如今日益丰富的数字多媒体编辑工具,为人们日常生活提供了便利的同时,也使对数字语音的篡改和伪造变得简单易行。被伪造的内容用在语音通讯,法庭举证和新闻报道等方面,势必会扭曲事实,破坏他人利益,严重影响数字证据的可信度和新闻媒体的公众认可度。对被攻击内容的篡改恢复,提高数字语音内容取证的可信度和法律效力,已成为多媒体信息安全的研究热点。基于数字语音信号在新闻报道和法庭举证环节的应用背景,以及数字语音信号真实性和完整性取证的实际需要,本文对数字语音为载体的取证和篡改恢复做了研究。具体工作如下:1、为了解决一些基于内容的数字语音取证技术存在的安全隐患,提出了一种抗特征分析替换攻击的语音内容取证算法。给出了语音信号能量比的定义,研究了基于信号能量比的水印嵌入方法。将帧号作为标识信息进行嵌入,在含水印信号被攻击后,通过帧号来定位被攻击的信号。该算法在对数字语音进行精确篡改定位的同时,提高了水印系统安全性。2、给出了基于离散小波变换DWT(Discrete Wavelet Transform)的数字语音压缩和重构方法,结合该方法提出了一种可恢复的数字语音取证算法。将压缩信号和帧号作为水印信息采用基于分块的方法嵌入到语音载体中。在语音信号遭受攻击后,用帧号来定位被攻击的位置,用压缩信号来恢复被攻击的内容。该算法在对数字语音进行取证的同时,能够恢复被攻击的内容,提高了取证结果的认可度和可靠性。
[Abstract]:With the development of digital signal processing technology and the popularization of high-speed network, digital signal has become the main carrier of information acquisition and exchange, such as binary digital speech signal. Digital speech signal has the advantages of easy storage, easy copy and easy transmission. It is widely used as information carrier in telephone communication, news report and so on. At the same time, it can also be used as electronic evidence in medical records and court evidence. Nowadays, the increasingly abundant digital multimedia editing tools not only provide convenience for people's daily life, but also make it easy to tamper with and forge digital speech. The falsified content used in voice communication, court evidence and news reports will certainly distort the facts, destroy the interests of others, and seriously affect the credibility of digital evidence and the public recognition of the news media. To recover the tampering of the attacked content and improve the credibility and legal effectiveness of digital voice content forensics has become a hot research topic of multimedia information security. Based on the application background of digital speech signal in news report and court evidence, as well as the actual need for authenticity and integrity of digital speech signal to obtain evidence, this paper studies the evidence collection and tampering recovery of digital speech as carrier. The specific work is as follows: 1. In order to solve the security problems of some content-based digital speech forensics techniques, a speech content forensics algorithm against feature analysis and replacement attacks is proposed. The definition of speech signal energy ratio is given, and the watermark embedding method based on signal energy ratio is studied. The frame number is embedded as the identification information, and after the watermark signal is attacked, the attacked signal is located by the frame number. The algorithm improves the security of watermarking system. It also improves the security of watermarking system. The method of digital speech compression and reconstruction based on discrete wavelet transform (DWT) is presented. Combined with this method, a recoverable digital speech forensics algorithm is proposed. The compressed signal and the frame number are embedded into the speech carrier using the block-based method. After speech signal is attacked, frame number is used to locate the location of the attack, and compressed signal is used to recover the attacked content. At the same time, the algorithm can recover the attacked content and improve the recognition and reliability of the result.
【学位授予单位】：信阳师范学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP309.7;TN912.3

【相似文献】