摘要
在音频文件里嵌入图像水印信息,极有可能引起音频质量的下降。该文首先利用Chebyshev混沌扩频原理对二值图像进行置乱,然后对音频信号进行小波变换,选取小波系数的低频部分来嵌入水印,再对小波系数进行均值量化,通过合理地确定量化步长从而实现在音频文件中嵌入图像水印信息。实验证实具有较好的鲁棒性和不可感知性。
Inserts the image watermarking information in the audio files, has the possibility to cause the audio frequency quality drop extremely, first use the Chebyshev chaotic spreading principle to binary images sets transforms randomly, Then carries on the wavelet transform to the audio signal, embed the low-frequency coefficient of watermarks, again carries on the quantification average to the wavelet coefficient, Thus through determines the quantification length of stride reasonably realization audio file watermarking information inserting, The experiment confirmed has the good robustness and not the perceptibility.
出处
《计算机安全》
2008年第9期22-25,29,共5页
Network & Computer Security
基金
河南省教育厅自然科学研究项目资助(2008B520047)
