中文摘要：

提出了一种基于音频特征的自适应盲音频水印算法：使用Logistic混沌二进制序列对水印进行加密，通过对音频帧的过零率及短时能量的分析，选取适当的阈值，确定嵌入水印的帧，并对选定音频帧整合分段后进行小波变换，同时采用能量动态调整自适应确定嵌入强度的方法，计算每段音频的近似分量的平均能量，结合基础嵌入强度，得音频帧中嵌入水印．实验结果表明：该算法具有较好的鲁棒性和不可感知性，具有一定的实用性．

英文摘要：

An adaptive rithm uses the Logistic blind audio watermarking algorithm based on audio features has been proposed. The algo- binary chaotic sequence of the watermark is encrypted, through the analysis and the short- time energy zero rate of the audio frame, select the appropriate threshold, is used to determine the embedding frame, and the wavelet transform of the sub audio frames selected after integration, method of simultaneous determination of embedding strength by dynamically adjusting the adaptive energy, average energy calculation of the approximate component of each audio, combining the embedding strength, embedding watermark in the audio frame. show that, the algorithm has good robustness, has a certain practicality.