论文部分内容阅读
针对多音源的音频信号,提出面向立体声的多源音频信息隐写算法.该算法将音源分离技术中的二进制时频掩码技术引入到信息隐写算法中,以各分离音源为载体,采用立体声混合嵌入的方式,通过修改时频域中奇数窗的部分系数,实现隐写信息的嵌入,该算法代替了传统信息隐写算法的单源嵌入方式.实验结果显示,该算法嵌入容量较大,并能有效抵抗常见信号处理攻击、StriMark Benchmark for Audio等攻击,平均误码率小于2.80%,算法具有很好的鲁棒性.
Aiming at the multi-source audio signal, this paper proposes a stereo-oriented multi-source audio information steganography algorithm.This algorithm introduces the binary time-frequency mask technology in audio source separation technology into the information steganography algorithm, uses the separated audio source as the carrier, Mixed embedding method to embed steganographic information by modifying partial coefficients of odd-numbered windows in the time-frequency domain, which replaces the single-source embedding method of traditional information steganography.The experimental results show that the embedding capacity of this algorithm is larger, And can effectively resist the common signal processing attacks, StriMark Benchmark for Audio attacks, the average bit error rate of less than 2.80%, the algorithm has good robustness.