基于小波域的均值量化数字音频水印算法*

提出了一种均值量化的小波域数字音频水印算法。均值量化可以减少攻击对水印的影响，算法选择合适的小波基对音频信号进行小波分解，对其中的中、低频系数进行均值量化，嵌入水印，水印提取过程不需要原始音频信号。实验结果表明，与一般水印量化算法相比，该算法对MP3、重量化、重采样、低通滤波和噪声干扰等信息处理型攻击具有更好的鲁棒性。

Mean Quantization Audio Watermarking Algorithm Based on Wavelet Transform

An audio watermarking algorithm based on mean quantization in the wavelet domain is presented. To reduce the effect of attacks on embedded watermark, mean quantization technique is used to embed the watermark. In this algorithm, the audio signal is decomposed using appropriate wavelet basis. Low frequency coefficients are selected to embed watermark using mean quantization method. The watermark can be extracted without the original digital audio signal. Experimental results show that compared with basic quantization watermarking algorithm, the algorithm presented in this paper is robust to many attacks, such as MP3 compression, requantization, resampling, low pass filtering, and noise addition and so on, and the robust performance has been improved highly.