Blind audio watermarking mechanism based on variational Bayesian learning

Xin TANG; Zhao-feng MA; Xin-xin NIU; Yi-xian YANG

doi:10.11959/j.issn.1000-436x.2015014

您当前的位置：

首页 >

文章列表页 >

Blind audio watermarking mechanism based on variational Bayesian learning

Academic paper | 更新时间：2024-06-05

- Blind audio watermarking mechanism based on variational Bayesian learning
- Journal on Communications Vol. 36, Issue 1, Pages: 121-128(2015)
- 作者机构：
  
  1. 北京邮电大学信息安全中心，北京100876
  2. 北京国泰信安科技有限公司，北京 100086
- 作者简介：
- 基金信息：
  
  The National Natural Science Foundation of China(60803157);The National Natural Science Foundation of China(90812001);The National Natural Science Foundation of China(61170271);The National Natural Science Foundation of China(61272519)
- DOI：10.11959/j.issn.1000-436x.2015014
  CLC： TP309
- Online First：2015-01，
  
  Published：25 January 2015
- 稿件说明：
移动端阅览
Xin TANG, Zhao-feng MA, Xin-xin NIU, et al. Blind audio watermarking mechanism based on variational Bayesian learning[J]. Journal on Communications, 2015, 36(1): 121-128.
DOI：

Xin TANG, Zhao-feng MA, Xin-xin NIU, et al. Blind audio watermarking mechanism based on variational Bayesian learning[J]. Journal on Communications, 2015, 36(1): 121-128. DOI： 10.11959/j.issn.1000-436x.2015014.

摘要

为了提高音频水印的检测性能，基于音频帧MFCC特征的统计特性，提出了一种音频水印盲检测方法。在音频帧的DCT系数上嵌入扩频水印，对嵌入水印的音频帧和原始音频帧分别提取MFCC特征进行训练，分别建立高斯混合模型，并通过变分贝叶斯学习方法估计出高斯混合模型的参数，检测时依据最大似然的原则。实验结果显示提出的方法在音频信号受到噪声干扰和恶意攻击的情况下，相对基于EM算法的方法在误检率上有明显降低，在小样本训练情况下具有更好的效果并且可以有效避兔过拟合的问题。

Abstract

In order to improve the performance of audio watermarking detection

a blind audio watermarking mechanism using the statistical characteristics based on MFCC features of audio frames was proposed.The spread spectrum watermarking was embedded in the DCT coefficients of audio frames.MFCC features extracted from watermarked audio frames as well as un-watermarked ones were trained to establish their Gaussian mixture models and to estimate the parameters by vatiational Bayesian learning method respectively.The watermarking was detected according to the maximum likelihood principle.The experimental results show that our method can lower the false detection rate compared with the method using EM algorithm when the audio signal was under noise and malicious attacks.Also

the experiments show that the proposed method achieves better performance in handling insufficient training data as well as getting rid of over-fitting problem.

关键词

Keywords

references

马兆丰，范科峰，陈铭等．支持时空约束的可信数字版权管理安全许可协议 [J ] ．通信学报， 2008 , 29 ( 10 ): 153 - 163 .

MA Z F , FAN K F , CHEN M , et al . Trusted digital rights management protocol supporting for time and space constraint [J ] . Journal on Communications , 2008 , 29 ( 10 ): 153 - 163 .

蒋铭，马兆丰，辛宇等．基于 DWT 和视觉加权的图像质量评价方法研究 [J ] ．通信学报， 2011 , 32 ( 9 ): 129 - 136 .

JIANG M , MA Z F , XIN Y , et al . Image quality evaluation method base on digital wavelet transform and vision weighted [J ] . Journal on Communications , 2011 , 32 ( 9 ): 129 - 136 .

冯涛，韩纪庆．双声道音频水印的同步及盲检测算法 [J ] ．通信学报， 2006 , 27 ( 10 ): 62 - 68 .

FENG T , HAN J Q . Synchronization and blind detect algorithm for dual channel audio watermark [J ] . Journal on Communications , 2006 , 27 ( 10 ): 62 - 68 .

孙中伟，朱岩，冯登国． DCT域图像水印的局部优化检测性能研究 [J ] ．电子学报， 2005 , 33 ( 5 ): 864 - 867 .

SUN Z W , ZHU Y , FENG D G . Performance analysis of DCT-domain watermark detection based on local optimum detection [J ] . Acta Electronica Sinica , 2005 , 33 ( 5 ): 864 - 867 .

AKHAEE M A , SAHRAEIAN S M E , MARVASTI F . Contourlet-based image watermarking using optimum detector in a noisy environment [J ] . IEEE Transactions on Image Processing , 2010 , 19 ( 4 ): 967 - 980 .

GUNSEL B , ULKER Y , ULKERULKER S . A statistical framework for audio watermark detection and decoding [A ] . Multimedia Content Representation,Classification and Security [C ] . Springer Berlin Heidelberg , 2006 . 241 - 248 .

林晓丹．基于高斯混合模型的 DCT 域水印检测方法 [J ] ．自动化学报， 2012 , 38 ( 9 ): 1445 - 1448 .

LIN X D . DCT-domain watermark detection using Gaussian mixture model [J ] . Acta Automatica Sinica , 2012 , 38 ( 9 ): 1445 - 1448 .

DEMPSTER A P , LAIRD N M , RUBIN D B . Maximum likelihood from incomplete data via the EM algorithm [J ] . Journal of the Royal Statistical Society.Series B (Methodological) , 1977 , 39 ( 1 ): 1 - 38 .

SHINOZAKI T , OSTENDORF M . Cross-validation EM training for robust parameter estimation [A ] . Proceedings of the 2007 IEEE International Conference on Acoustics,Speech and Signal,ICASSP 2007 [C ] . 2007 .IV-437-IV-440.

MOATTAR M H , HOMAYOUNPOUR M M . Text-independent speaker verification using variational Gaussian mixture model [J ] . ETRI Journal , 2011 , 33 ( 6 ): 914 - 923 .

CORDUNEANU A , BISHOP C M . Variational bayesian model selection for mixture distributions [A ] . Artificial Intelligence and Statistics [C ] . Waltham,MA : Morgan Kaufmann 2001 . 27 - 34 .

COX I J , KILIAN J , LEIGHTON F T , et al . Secure spread spectrum watermarking for multimedia [J ] . IEEE Transactions on Image Processing , 1997 , 6 ( 12 ): 1673 - 1687 .

KIROVSKI D , MALVAR H S . Spread-spectrum watermarking of audio signals [J ] . IEEE Transactions on Signal Processing , 2003 , 51 ( 4 ): 1020 - 1033 .

MALIK H , KHOKHAR A , ANSARI R . Improved watermark detection for spread-spectrum based watermarking using independent component analysis [A ] . Proceedings of the 5th ACM Workshop on Digital Rights Management [C ] .ACM 2005 . 102 - 111 .

BISHOP C M , NASRABADI N M . Pattern Recognition and Machine Learning [M ] . New York : Springer , 2006 .

WEYCHAN R , MARCINIAK T . Analysis of differences between MFCC after multiple GSM transcodings [J ] . Przegląd Elektrotechniczny Selected Full Texts , 2012 , 88 ( 6 ): 24 - 29 .

BESSON O , BIDON S , TOURNERET J Y . Covariance matrix estimation with heterogeneous samples [J ] . IEEE Transactions on Signal Processing , 2008 , 56 ( 3 ): 909 - 920 .

GÖRÜR D , RASMUSSEN C E . Dirichlet process Gaussian mixture models:choice of the base distribution [J ] . Journal of Computer Science and Technology , 2010 , 25 ( 4 ): 653 - 664 .

MURPHY K P . Conjugate Bayesian Analysis of the Gaussian Distribution [R ] . Technical Report,UBC , 2007 .

DO M N , VETTERLI M . Wavelet-based texture retrieval using generalized Gaussian density and Kullback-Leibler distance [J ] . IEEE Transactions on Image Processing , 2002 , 11 ( 2 ): 146 - 158 .

SHRIBERG E , FERRER L , KAJAREKAR S , et al . Modeling prosodic feature sequences for speaker recognition [J ] . Speech Communication , 2005 , 46 ( 3 ): 455 - 472 .

STEINEBACH M , PETITCOLAS F A P , RAYNAL F , et al . StirMark benchmark:audio watermarking attacks [A ] . Proceedings of the 2011 International Conference on Information Technology:Coding and Computing [C ] . 2001 . 49 - 54 .

Views

2278

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Semi-supervised signal modulation recognition method based on dual-confidence fusion mechanism

Incremental clustering method based on Gaussian mixture model to identify malware family

GMM-based localization algorithm under NLOS conditions

Statistical covariance blind detection algorithm based on cholesky factorization in cognitive radio network

Synchronization and blind detect algorithm for dual channel audio watermark

Related Author

QIU Xin

LI Jinhai

SONG Liubin

YANG Chaosan

MIAO Xuchen

LIU Wenxue

Jianwei HU

Xin CHE

Related Institution

School of Integrated Circuits, University of Chinese Academy of Sciences

Communications and Information Engineering R&D Center IMECAS

School of Network and Information Security, Xidian University

Institute of Cyberspace Security,Huazhong University of Science and Technology

College of Information Science and Engineering,Northeastern University

AI问答

⁰