CAJ | 학술논문

针对数字音频产品版权保护问题,设计了一种基于混沌加密和重复码调制的复倒谱变换数字音频水印算法.算法把音频信号划分为包含相同采样点的若干帧,对每一帧实施复倒谱变换,将一幅包含版权信息的二值图像作为水印,利用混沌序列对其加密,再利用重复码进行纠错后嵌入到原始音频的复倒谱系数中,水印提取时不需要原始音频信号.实现了水印的盲检测.仿真实验证明,算法具有较强的鲁棒性和不可感知性.能经受添加噪声、低通滤波、重采样、重新量化、音频格式转换等常见信号的处理及攻击.
침대수자음빈산품판권보호문제,설계료일충기우혼돈가밀화중복마조제적복도보변환수자음빈수인산법.산법파음빈신호화분위포함상동채양점적약간정,대매일정실시복도보변환,장일폭포함판권신식적이치도상작위수인,이용혼돈서렬대기가밀,재이용중복마진행규착후감입도원시음빈적복도보계수중,수인제취시불수요원시음빈신호.실현료수인적맹검측.방진실험증명,산법구유교강적로봉성화불가감지성.능경수첨가조성、저통려파、중채양、중신양화、음빈격식전환등상견신호적처리급공격.
For having the copyright security of the digital audio frequency product,this article proposes a blind digital audio watermarking algorithm in complex cepstrum domain based on the chaotic encrypting and repeatencode modulating.The audio signal is divided into several frames which include the same samples and the frame is transformed with complex cepstnnn.Then taking a binary image which having the message of copyright as watermark,the chaotic sequence encrypts it and the repeatencode can be embedded into the complex cepstrum coefficients after rectifying error while the watermark can be extracted without the original audio signal.By doing this,blind detection of the watermark is achieved.Experimental result demonstrates that the algorithm is of much robustness and imperceptibility and thus can endure processing and attacks from common signals,such as resampling,extra-noise,low-pass filter,remodulation and audio frequency format.conversion etc.