摘要
高度鲁棒数字音频水印算法研究是一项富有挑战性的工作。为了在数字音频水印技术中提高抗去同步攻击,文章以能量均值、特征点为理论基础,提出了一种新颖的鲁棒性数字音频水印算法。该算法选取数字音频的局部能量峰值点作为特征,提取稳定的特征点;以特征点为标记,对特征点后的区域进行DWT,然后对低频系数分段,利用能量均值和音频幅值成正比的方法,将水印嵌入到原始音频的小波域中。水印提取时无需原始图像。仿真实验表明,所提出的算法对常规信号处理和去同步攻击均具有较好的鲁棒性。
A robust digital audio watermarking algorithm based on the feature points and energy mean value is proposed in this paper.Firstly,to choose the digital audio local energy peaks as feature to extract the stable feature points;secondly,to mark the feature points and segment the audio signal behind the feature points;then,embed the watermark bits into wavelet domain by energy mean value.Meanwhile,the algorithm can extract the watermark without the help from the original digital audio signals.Simulation results show that the proposed watermarking scheme is not only robust against common signal processing,but also robust against the desynchronization attacks.
出处
《中国体视学与图像分析》
2011年第2期124-130,共7页
Chinese Journal of Stereology and Image Analysis
关键词
音频水印
特征点
能量均值
鲁棒性
audio watermarking
feature points
energy mean value
robustness
