摘要
提出了一种新的自适应音频数字水印算法.它首先将视觉上可辨的二值水印图像降为一维序列,再将水印序列与m序列做扩频调制.最后对语音信号进行分段离散余弦变换,把经过调制的水印信息自适应地嵌入到变换域绝对值最大的系数上.仿真实验证明了算法的稳健性和不可觉察性.
A novel adaptive digital watermark algorithm has been proposed to first convert the visually recognizable binary watermarking image into the onedimension sequence, then to modulate the sequence spectrumspread with msequence,and finally,divide the audio signal into segments and embed the modulated watermark into the biggest coefficient of the transform domain adaptively. Experimental results demonstrate the algorithm is robust and imperceptible.
出处
《哈尔滨工程大学学报》
EI
CAS
CSCD
2003年第3期317-320,共4页
Journal of Harbin Engineering University
基金
哈尔滨市后备学科带头人基金(0171218025)
全国优秀博士学位论文作者专项基金(200037)
教育部优秀青年教师资助计划项目
及高等学校优秀青年教师教学科研奖励计划资助.
关键词
数字水印
自适应
鲁棒性
不可觉察性
digital watermark
adaptation
robustness
imperceptibility
