一种基于DWT-QR的脆弱性音频水印算法  

A Fragile Audio Watermarking Algorithm Based on DWT-QR

作　　者：杨志疆[1] 叶阿勇[2] 

机构地区：[1]漳州职业技术学院计算机工程系,福建漳州363000 [2]福建师范大学网络安全与密码技术重点实验室,福建福州350007

出　　处：《吉林师范大学学报（自然科学版）》2015年第1期130-135,共6页Journal of Jilin Normal University:Natural Science Edition

基　　金：国家自然科学基金项目(61202452);福建省中青年教师教育科研项目(JA13386)

摘　　要：为了更好地实现音频的内容认证,提出了一种基于DWT-QR脆弱音频水印算法.算法首先对音频分帧,然后在每一帧前部分嵌入的同步码保证水印的嵌入与检测帧同步,并在帧后部分基于DWT-QR变换把脆弱水印嵌入到高频分量中.在水印嵌入前,采用二次混沌加密的方法为水印信息进行加密处理,保证水印图像的安全性;水印采用自适应均值量化的方法来嵌入,实现了盲检测.仿真实验表明,该算法简单、高效、透明性与易碎性强,而且能较准确地实现对恶意篡改的定位与统计.In order to better realize the content authentication of audio,a fragile watermarking algorithm was proposed based on DW-QR. Firstly,the audio was divided into frames,then the synchronization code was embedded in the first part of each frame to ensure the frame synchronization both embedding and detection,and the fragile watermark was embedded in high frequency components of DWT-QR. Chaotic encryption used twice to ensure the security of watermark image before the watermark embedding. The watermark was embedded by the adaptive mean quantization steps,realized the blind detection. Simulation results showed that the algorithm was very simple,efficient,transparent,fragile,and accurate localization and statistics for tamper.

关 键 词：帧同步 DWT-QR 脆弱水印 内容认证 篡改定位 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]