论文部分内容阅读
为了使数字水印实现多种功能,提出了一种能在同一音频信号中同时嵌入多个水印的新方法.首先,载体音频信号被分隔成指定长度的帧,并且所有成员水印被编码成一个混合水印.然后,利用回声隐藏技术将混合水印中的二进制比特嵌入到音频信号帧中.水印提取不需要原始音频信号.为了提高水印算法的提取正确率和抵御常规信号处理的鲁棒性,功率倒谱的自相关被用于检测携带水印的音频信号帧中的回声延时以提取出混合水印,并且相应的解码算法被用于分离成员水印.计算机仿真结果表明所提出的水印算法对常规的信号处理,如Mp3压缩、重采样、重量化、低通滤波以及白噪声具有很强的鲁棒性.
In order to enable digital watermarking to realize a variety of functions, a new method of embedding multiple watermarks simultaneously in the same audio signal is proposed.Firstly, the carrier audio signal is divided into frames of a specified length, and all the member watermarks are encoded into a mixed Watermarking.Then, the binary bit in the hybrid watermark is embedded into the audio signal frame by using the echo hiding technique.Watermark extraction does not need the original audio signal.In order to improve the extraction accuracy of the watermarking algorithm and resist the robustness of the conventional signal processing, the power down The autocorrelation of the spectrum is used to detect the echo delay in the watermarked audio signal frame to extract the hybrid watermark, and the corresponding decoding algorithm is used to separate the member watermarks.The computer simulation results show that the proposed watermarking algorithm performs better than the traditional signal Processing, such as Mp3 compression, resampling, weighting, low-pass filtering and white noise is robust.