基于小波与复倒谱变换的音频数字水印算法  

Audio Frequency Digital Watermark Algorithm Based on Wavelet and Complex Cepstrum Transform

作　　者：刘素楠[1] 谢中梅[1] 胡春生[1] 

机构地区：[1]江西应用技术职业学院信息工程系,赣州341000

出　　处：《计算机工程》2009年第23期152-154,共3页Computer Engineering

基　　金：江西省教育厅科研基金资助项目(2007208)

摘　　要：为提高水印安全性,提出一种基于小波分解和复倒谱变换的音频数字水印算法。通过对原始语音信号进行三级小波分解,提取小波近似分量,对小波近似分量进行复倒谱变换。实验结果证明,该算法具有较强不可感知性和鲁棒性,能经受添加噪声、低通滤波、重采样、有损压缩等常见信号攻击。To improve the security of watermark, this paper proposes an audio digital frequency watermark algorithm based on wavelet transform and complex cepstrum transform. This algorithm decomposes original audio signal in three level wavelet, extracts the wavelet approximate components, and transforms the wavelet approximate components in complex cepstrum domain. Experimental results show that the algorithm has high imperceptibility and robustness and can endure attacks from common signals, such as extra-noise, low-pass filter, resampling and lossy compression.

关 键 词：数字水印 小波变换 复倒谱 混沌 重复码 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]