基于混合分类器的AMR-WB语音隐写分析方法
发布时间：2021-12-16 19:41
基于自适应多速率宽带语音编码的信息隐藏方法具有较强的隐蔽性和抗检测能力。为检测导谱频率矢量索引搜索中的量化索引调制,提出了一种基于校准技术和混合分类器的隐写分析方法。其中,一种基于长短期记忆模型的特征提取方法被用于提取导谱频率的关联性特征,导谱频率量化索引的概率分布则作为导谱频率的长时分布特征。以上两种特征通过校准技术处理后,被送入支持向量机中进行训练,可获得具有较高分类性能的分类器。实验结果表明,在不同嵌入率和语音长度下,所提方法均有较高的分类正确率。
【文章来源】：通信技术. 2020,53(10)
【文章页数】：7 页
【参考文献】：
期刊论文
[1]Steganography in low bit-rate speech streams based on quantization index modulation controlled by keys[J]. HUANG YongFeng,TAO HuaiZhou,XIAO Bo,CHANG ChinChen. Science China（Technological Sciences）. 2017(10)
[2]Detection of quantization index modulation steganography in G.723.1 bit stream based on quantization index sequence analysis[J]. Song-bin LI 1,2,Huai-zhou TAO1,2,Yong-feng HUANG 1,2(1Department of Electronic Engineering,Tsinghua University,Beijing 100084,China)(2Tsinghua National Laboratory for Information Science and Technology,Beijing 100084,China). Journal of Zhejiang University-Science C（Computers & Electronics）. 2012(08)
本文编号：3538717
【文章来源】：通信技术. 2020,53(10)
【文章页数】：7 页
【参考文献】：
期刊论文
[1]Steganography in low bit-rate speech streams based on quantization index modulation controlled by keys[J]. HUANG YongFeng,TAO HuaiZhou,XIAO Bo,CHANG ChinChen. Science China（Technological Sciences）. 2017(10)
[2]Detection of quantization index modulation steganography in G.723.1 bit stream based on quantization index sequence analysis[J]. Song-bin LI 1,2,Huai-zhou TAO1,2,Yong-feng HUANG 1,2(1Department of Electronic Engineering,Tsinghua University,Beijing 100084,China)(2Tsinghua National Laboratory for Information Science and Technology,Beijing 100084,China). Journal of Zhejiang University-Science C（Computers & Electronics）. 2012(08)
本文编号：3538717
本文链接：https://www.wllwen.com/kejilunwen/wltx/3538717.html
