一种抗剪切的DWT域多重音频数字水印算法

A multiple audio watermarking algorithm based on DWT against cropping

提出了一种能够抵抗剪切攻击的鲁棒性DWT域多重音频数字水印算法。该算法结合音频分帧技术及小波变换理论,将置乱、降维处理后的不同彩色数字图像信息嵌入到宿主音频信号的低频小波系数中,实现了多重彩色图像水印在宿主音频中的嵌入与提取。实验结果表明,该算法不仅能够较好地抵抗常规攻击,对剪切攻击也具有较强的鲁棒性,即使是在受到强剪切攻击后仍然能够提取出较清晰的(多重)水印信号。

A robust muhiple audio watermarking algorithm based on DWT against cropping was proposed. The algorithm combines the audio sub-frame technology and the DWT theory embedded in the host audio signal in the low-frequency wavelet coefficients to different color images after scrambling and dimensional. The algorithm realization of a multi-color image in the host audio watermark embedding and extraction. The experimental results show that the multiple audio watermarking algorithm, in addition to better resistance to conventional attacks, attacks on the shear is relatively robust, even in the post-attack by the strong shear will still be able to extract a clearer definition of （multiple） watermark signal.