一种基于卷积码的小波域音频水印盲检算法

An Audio Watermarking Blind Algorithm Based on Convolution Codes in Wavelet Domain

水印系统的顽健性是数字水印研究者最为关注的问题。借助卷积码的编解码思想,在小波域研究了自适应于音频的扩频水印算法,将视觉可辨的水印经过预处理后嵌在所选的中、低频小波子带。为了提高算法的顽健性,提出了适合Viterbi软解码的水印检测改进算法,水印检测不需要原始音频。实验结果表明,基于卷积码的音频扩频水印算法具有很高的顽健性,能完全抵制信号处理型攻击。

The robustness of watermarking system is more and more paid attention by digital watermarking researcher. With the help of convolutional idea, the adaptive audio watermarking algorithm is investigated in wavelet domain in this paper. The visually recognizable watermark pre-processed is embedded into middle and low frequency sub-band in wavelet domain. In order to enhance robustness, the detecting improved algorithm which use Viterbi soft-decision is presented, and detecting watermark don’t need original audio. The experimental results show that the proposed algorithm is provided with high robustness, and sustainable against compression algorithms, as well as common signal processing manipulation attacks.