摘要
以往基于离散小波变换的音频数字水印算法鲁棒性不强,无法对数字音频信号的版权进行真正地保护,因此改进了算法,利用心理声学模型,控制水印信号的嵌入,通过计算音频信号的掩蔽阀值,在保证水印嵌入不可察觉的前提下,增大水印的嵌入强度。仿真实验证明,该算法具有较强的鲁棒性和不可感知性,能有效抵御各种常见攻击,并且能实现盲检测。
Because the robustness of the old audio digital watermarking algorithm based on DWT transform is not strong enough to offer efficacious copyright protection for digital audio signals, this paper presents an improved algorithm. Using psychoacoustic model and calculating the masking threshold of the audio signal, this improved algorithm can control embedding the watermark signal and increase embedding strength in an undiscoverable way. The experimental results show that this watermarking algorithm has good robustness and imperceptibility. It can resist various usua.] attacks and implement blind detection.
出处
《计算机安全》
2008年第8期8-12,共5页
Network & Computer Security
关键词
心理声学模型
小波变换
音频数字水印
盲检测
鲁棒性
Psychoacoustic model
Wavelet transform
Audio digital watermarking
Blind detection
Robustness
