电子学报 ›› 2015, Vol. 43 ›› Issue (8): 1513-1517.DOI: 10.3969/j.issn.0372-2112.2015.08.007

• 学术论文 • 上一篇    下一篇

基于共生特性的G.729A基音调制信息隐藏的检测

贾已真2, 李松斌1, 蒋雨欣1, 戴琼兴1, 邓浩江3   

  1. 1. 中国科学院声学研究所南海研究站, 海南海口, 570105;
    2. 海南大学信息科学技术学院, 海南海口 570228;
    3. 国家网络新媒体工程技术研究中心, 北京 100190
  • 收稿日期:2014-04-23 修回日期:2014-07-28 出版日期:2015-08-25 发布日期:2014-12-05
  • 通讯作者: 李松斌
  • 作者简介:贾已真 女,1989年出生,河南商丘人.2012年毕业于海南大学信息科学技术学院,获得学士学位.现为海南大学硕士研究生,研究方向为网络与流媒体技术. E-mail: JYZ231@126.com
  • 基金资助:

    国家自然科学基金(No.61303249);海南省自然科学基金(No.614236);海南省重大科技项目课题(No.JDJS2013006)

G.729A Pitch Modulation Information Hiding Detection Based on Symbiotic Characteristics

JIA Yi-zhen2, LI Song-bin1, JIANG Yu-xin1, DAI Qiong-xing1, DENG Hao-jiang3   

  1. 1. Haikou Laboratory, Institute of Acoustics, Chinese Academy of Sciences, Haikou, Hainan 570105, China;
    2. College of Information Science and Technology, Hainan University, Haikou, Hainan 570228, China;
    3. National Network New Media Engineering Research Center, Beijing 100190, China
  • Received:2014-04-23 Revised:2014-07-28 Online:2015-08-25 Published:2014-12-05

摘要:

提出了一种G.729A自适应码本分组基音调制信息隐藏的检测算法.对语音码流的分析发现,通过基音预测进行信息隐藏将改变相邻语音帧中基音周期估计值的共生特性.通过量化这种共生特性,并经过PCA(Principal Component Analysis,主成分分析)降维获得对隐写检测敏感的特征向量.最后基于特征向量和SVM (Support Vector Machine,支持向量机)构建隐写检测器.对不同语音样本数据集的检测表明,当语音长度在2s及以上时,检测正确率均超过96%.此检测算法是一种有效的压缩域信息隐藏检测方法.

关键词: 基音调制信息隐藏, 共生特性, 隐写检测, 主成分分析, 支持向量机

Abstract:

A detection algorithm of pitch modulation information hiding in G.729A low bit-rate speech codec is proposed.The analysis of speech stream showed that pith modulation information hiding would change the pitch symbiotic characteristics of adjacent speech frames.We designed a model to quantify these pitch symbiotic characteristics for steganalysis.However, the dimension of quantitative feature vector of pitch symbiotic characteristics was too high, so PCA (Principal Component Analysis) was employed to reduce the dimension of the feature vector.Finally, we built a pitch modulation information hiding detector based on the dimension reduced feature vector and SVM (Support Vector Machine) classifier.Experiments on different speech datasets show that the proposed steganalysis algorithm is very effective: the accuracy is more than 96% when speech length equals to or is greater than 2s.So this paper gives an effective method for compression domain information hiding detection.

Key words: pitch modulation information hiding, symbiotic characteristics, steganography detection, principal component analysis (PCA), support vector machine (SVM)

中图分类号: