一种鲁棒音频数字水印时域算法

摘要

A robust time-domain audio digital watermark algorithm based on energy difference comparison is proposed. The audio carrier is divided into several sections with equal length according to the watermark sequence being to be embedded,and each section is divided into several segments with equal length. The watermark sequence is embedded into the audio carrier by comparing and modifying the energy difference of the first two segments and the last two segments of each section,and combinating the time-domain masking effect of human auditory system and the watermark sequence being to be embedded. The simulation re-sults show that the algorithm has good transparency and good robustness against various audio attacks,and can realize blind de-tection.%提出了一种强鲁棒性的基于能量差值比较的时域音频数字水印算法。根据要嵌入的水印序列将音频载体分为长度相等的若干节，每节再分为长度相等的若干段，通过比较每节前两段和后两段的能量差值，结合要嵌入的水印序列和人类听觉系统的时域掩蔽效应，采取修改每节前两段或后两段能量差值的方法将水印序列嵌入到音频载体中。实验结果表明，该算法透明性良好，对多种音频攻击均具有良好的鲁棒性，并可实现盲检测。