摘要
数字音频篡改被动检测是指不依赖任何预先嵌入的信息来鉴别数字音频真伪的技术,其最主要研究内容是判定数字音频的真实性和完整性,在司法取证、新闻公正、知识产权保护等领域有着广泛的应用前景。目前领域内相关综述主要从数字音频主动、被动取证总体框架开展,并未专门针对数字音频篡改被动取证研究进行系统全面总结,且涉及被动取证部分存在时效性不足的问题。据此首先总结了数字音频篡改被动检测的任务模型和取证框架,接着依据篡改手段、检测策略、所使用的统计特征及模型,将目前的数字音频篡改被动检测方法分为四类:基于篡改操作的检测方法、基于数字音频重压缩的检测方法、基于录音设备和音频录制环境的检测方法、基于数字音频信号自身统计特性的检测方法,然后分析了每种方法所采用的典型算法和扩展手段,并对不同检测算法进行性能比较,然后对这四类方法的检测特点和使用范围进行总结。最后综合近年来国内外研究人员的主要成果,总结了数字音频篡改被动检测研究面临的问题和挑战,并对未来的研究进行了展望。
Passive detection of digital audio tampering refers to the technology that does not rely on any pre-embedded information to identify the authenticity of digital audio. The main research content is to determine the authenticity and integrity of digital audio. It has a wide application prospect in the fields of court forensics, news justice, intellectual property protection and so on. At present, the related reviews in the field are mainly from the point of overall framework of active and passive digital audio forensics, and they are without a systematic and comprehensive summary of passive detection of digital audio tampering. What's more, the passive audio forensics research involving in the above reviews is with the problem of inadequate timeliness. This paper first summarizes the task model and forensics framework of digital audio tamper detection, and then divides the present digital audio tampering passive detection methods into four categories according to the tamper means, detection strategies, statistical characteristics and models used:the detection method based on tamper operation, the detection method based on the digital audio recompression, the detection method based on the recording equipment and the audio recording environment, and the detection method based on the statistical characteristics of the digital audio signal. Then it analyzes the typical algorithm and the extended means of each method, and compares the performance of different detection algorithms, then four kinds of methods for the detection of the characteristics and application scope are summarized. Finally, the main research achievements of researchers at home and abroad in digital audio tampering passive detection technology are summarized, the problems and challenges of digital audio tampering passive detection technology and the prospect of future research are put forward.
引文
[1]Zakariah M,Khan M K,Malik H.Digital multimedia audio forensics:past,present and future[J].Multimedia Tools and Applications,2017,77(1):1009-1040.
[2]胡永健.数字多媒体取证技术综述[J].计算机应用,2010,30(3):657-662.
[3]包永强,梁瑞宇,丛韫,等.音频取证若干关键技术研究进展[J].数据采集与处理,2016(2):252-259.
[4]钟巍.音频取证综述[C]//全国信息隐藏暨多媒体信息安全学术大会,2010.
[5]Gupta S,Cho S,Kuo C C J.Current developments and future trends in audio authentication[J].IEEE Multimedia,2011,19(1):50-59.
[6]Zimba M,Sun X.DWT-PCA(EVD)based copy-move image forgery detection[J].International Journal of Digital Content Technology&Its Applications,2011,5(1):251-258.
[7]Liu F,Feng H.An efficient algorithm for image copymove forgery detection based on DWT and SVD[J].International Journal of Security&Its Applications,2014,8(5):377-390.
[8]Luo D,Korus P,Huang J.Band energy difference for source attribution in audio forensics[J].IEEE Transactions on Information Forensics&Security,2018,PP(99):1.
[9]Luo W,Li H,Yan Q,et al.Improved audio steganalytic feature and its applications in audio forensics[J].ACMTransactions on Multimedia Computing,Communications,and Applications(TOMM),2018,14(2):43.
[10]Hua G,Bi G,Thing L.On practical issues of ENF based audio forensics[J].IEEE Access,2017:1-10.
[11]Shi Qian,Ma Xiaohong.Detection of audio interpolation based on singular value decomposition[C]//International Conference on Awareness Science and Technology,2011:287-290.
[12]刘潘梅,孙容海,吴建源.一种新的区域复制图像篡改盲检测技术[J].计算机工程与应用,2012,48(9):161-164.
[13]Imran M,Ali Z,Bakhsh S T,et al.Blind detection of copy-move forgery in digital audio forensics[J].IEEEAccess,2017,PP(99):1.
[14]姚秋明,柴佩琪.基于期望最大化算法的音频取证中的篡改检测[J].计算机应用,2006,26(11):2598-2601.
[15]陈雁翔,吴玺.音频盲取证中一种基于EM的重采样检测方法[J].电路与系统学报,2012,17(4):118-123.
[16]Popescu A C,Farid H.Exposing digital forgeries by detecting traces of resampling[J].IEEE Transactions on Signal Processing,2005,53(2):758-767.
[17]侯丽敏,吴威麒.基于二阶差分的音频重采样盲取证[J].上海大学学报,2014,20(3):304-312.
[18]Mahdian B,Saic S.Blind authentication using periodic properties of interpolation[J].IEEE Transactions on Information Forensics&Security,2008,3(3):529-538.
[19]Ding Qi,Wang Xu,Ping Xijian.Audio interpolation and stitching detection based on band-partitioning spectral smoothness[C]//WASE International Conference on Information Engineering,2009,1:189-192.
[20]Yang R,Shi Y Q,Huang J.Detecting double compression of audio signal[C]//IS&T/SPIE Electronic Imaging.[S.l.]:International Society for Optics and Photonics,2010.
[21]Tao B,Wang R,Yan D,et al.Audio tampering detection based on quantization artifacts[C]//International Conference on Cloud Computing and Security.Cham:Springer,2016:430-439.
[22]柳永娟,马晓红.2010通信理论与信号处理学术年会论文集[C].北京:电子工业出版社,2010:525-530.
[23]Yang R,Shi Y Q,Huang J.Defeating fake-quality MP3[C]//Proceedings of the 11th ACM Workshop on Multimedia and Security,2009:117-124.
[24]D’Alessandro B,Shi Y Q.MP3 bit rate quality detection through frequency spectrum analysis[C]//Proceedings of the 11th ACM Workshop on Multimedia and Security,2009:57-62.
[25]Liu Q,Sung A H,Qiao M.Detection of double MP3compression[J].Cognitive Computation,2010,2(4):291-296.
[26]Qiao M,Sung A H,Liu Q.Revealing real quality of double compressed MP3 audio[C]//Proceedings of the18th ACM International Conference on Multimedia,2010:1011-1014.
[27]Korycki R.Authenticity examination of compressed audio recordings using detection of multiple compression and encoders’identification[J].Forensic Science International,2014,238:33-46.
[28]Yang R,Qu Z,Huang J.Detecting digital audio forgeries by checking frame offsets[C]//Proceedings of the 10th ACM Workshop on Multimedia and Security,2008:21-26.
[29]Yang R,Qu Z,Huang J.Exposing MP3 audio forgeries using frame offsets[J].ACM Transactions on Multimedia Computing,Communications,and Applications(TOMM),2012,8(2S):35.
[30]杨婧,范梦迪.一种改进的MP3被动篡改定位检测算法[J].信息网络安全,2014(10):7-10.
[31]Yan D,Wang R,Zhou J,et al.Compression history detection for MP3 audio[J].KSII Transactions on Internet&Information Systems,2018,12(2):662-675.
[32]Zhou J,Wang R,Jin C,et al.Multiple MP3 compression detection based on the statistical properties of scale factors[C]//International Workshop on Digital Watermarking.[S.l.]:Springer International Publishing,2015:51-60.
[33]Shen Y,Jia J,Cai L,et al.Detecting double compressed AMR-format audio recordings[C]//第十届中国语音学学术会议论文集,2012:1-5.
[34]Luo D,Yang R,Huang J,et al.Detecting double compressed AMR audio using deep learning[C]//2014 IEEEInternational Conference on Acoustics,Speech and Signal Processing(ICASSP),Florence,Italy,2014:2669-2673.
[35]Luo D,Yang R,Huang J.Identification of AMR decompressed audio[J].Digital Signal Processing,2015,37:85-91.
[36]Luo D,Yang R,Li B,et al.Detection of double compressed AMR audio using stacked autoencoder[J].IEEETransactions on Information Forensics&Security,2017,PP(99):1.
[37]黄其娟,王让定,严迪群.基于比例因子转移概率的AAC音频压缩历史检测算法[J].电信科学,2017,33(11):93-101.
[38]Jin C,Wang R,Yan D,et al.An efficient algorithm for double compressed AAC audio detection[J].Multimedia Tools&Applications,2016,75(8):1-18.
[39]Grigoras C.Digital audio recording analysis:the electric network frequency criterion[J].International Journal of Speech Language and the Law,2005,12(1):63-76.
[40]Grigoras C.Applications of ENF criterion in forensic audio,video,computer and telecommunication analysis[J].Forensic Science International,2007,167(2):136-145.
[41]Liu Y,Yao C,Sun C,et al.The authentication of digital audio recordings using power system frequency[J].IEEE Potentials,2014,33(2):39-42.
[42]Dosiek L.Extracting electrical network frequency from digital recordings using frequency demodulation[J].IEEESignal Processing Letters,2015,22(6):691-695.
[43]Rodríguez N,Nicolalde P,Apolinario A.Evaluating digital audio authenticity with spectral distances and ENF phase change[C]//IEEE International Conference on Acoustics,Speech and Signal Processing,2009:1417-1420.
[44]Rodríguez N,Apolinário A,Biscainho P.Audio authenticity:detecting ENF discontinuity with high precision phase analysis[J].IEEE Transactions on Information Forensics and Security,2010,5(3):534-543.
[45]Hu Y,Li T,Lv Z,et al.Audio forgery detection based on max offsets for cross correlation between ENF and reference signal[C]//International Workshop on Digital Watermarking.Berlin,Heidelberg:Springer,2012:253-266.
[46]Lv Z,Hu Y,Li T,et al.A blind audio forgery detection algorithm based on dual judgment mechanisms[J].Journal of South China University of Technology,2014,42(8):70-76.
[47]Esquef P A A,Apolinário J A,Biscainho L W P.Edit detection in speech recordings via instantaneous electric network frequency variations[J].IEEE Transactions on Information Forensics and Security,2014,9(12):2314-2326.
[48]Reis P,Miranda R,Galdo G.ESPRIT-Hilbert-based audio tampering detection with SVM classifier for forensic analysis via electrical network frequency[J].IEEE Transactions on Information Forensics&Security,2017,12(4):853-864.
[49]何朝霞,潘平,罗辉.复制粘贴音频信号的篡改检测技术研究[J].中国测试,2016,42(7):107-111.
[50]Garg R,Varna A L,Wu M.Modeling and analysis of electric network frequency signal for timestamp verification[C]//2012 IEEE International Workshop on Information Forensics and Security(WIFS),2012:67-72.
[51]Lin X,Kang X.Supervised audio tampering detection using an autoregressive model[C]//IEEE International Conference on Acoustics,Speech and Signal Processing,2017:2142-2146.
[52]Guang Hua,Zhang Y,Goh J,et al.Audio authentication by exploring the absolute-error-map of ENF signals[J].IEEE Transactions on Information Forensics and Security,2016,11(5):1003-1016.
[53]Lin X,Liu J,Kang X.Audio recapture detection with convolutional neural networks[J].IEEE Transactions on Multimedia,2016,18(8):1480-1487.
[54]Esquef P,Apolinario J,Biscainho L.Improved edit detection in speech via ENF patterns[C]//IEEE International Workshop on Information Forensics and Security,2015:1-6.
[55]Rodríguez N,Apolinario A,Biscainho P.Audio authenticity based on the discontinuity of ENF higher harmonics[C]//IEEE Signal Processing Conference,2010:1-5.
[56]Chuang W H,Garg R,Wu M.Anti-forensics and countermeasures of electrical network frequency analysis[J].IEEE Transactions on Information Forensics and Security,2013,8(12):2073-2088.
[57]Savari M,Wahab A W A,Anuar N B.High-performance combination method of electric network frequency and phase for audio forgery detection in battery-powered devices[J].Forensic Science International,2016,266:427-439.
[58]Kraetzer C,Oermann A,Dittmann J,et al.Digital audio forensics:a first practical evaluation on microphone and environment classification[C]//Proceedings of the 9th Workshop on Multimedia&Security,2007:63-74.
[59]邵松年,黄征,等.数字音频与录制设备的相关性研究[J].计算机工程,2009,35(19):224-250.
[60]Buchholz R,Kraetzer C,Dittmann J.Microphone classification using Fourier coefficients[C]//International Workshop on Information Hiding.Berlin,Heidelberg:Springer,2009:235-246.
[61]Cuccovillo L,Mann S,Tagliasacchi M,et al.Audio tampering detection via microphone classification[C]//2013IEEE 15th International Workshop on Multimedia Signal Processing(MMSP 13),Pula,Sardinia,Italy,2013:177-182.
[62]Koenig B E,Lacey D S.The average direct current offset values for small digital audio recorders in an acoustically consistent environment[J].Journal of Forensic Sciences,2014,59(4):960-966.
[63]Muhammad G,Alghathbar K.Environment recognition for digital audio forensics using MPEG-7 and melcepstralfeatures[J].Journal of Electrical Engineering,2011,62(4):199-205.
[64]Malik H,Farid H.Audio forensics from acoustic reverberation[C]//2010 IEEE International Conference on Acoustics,Speech,and Signal Processing(ICASSP),2010:1710-1713.
[65]何少岩,陈蕉容.基于录制环境检测的数字音频取证研究[J].计算机工程与设计,2013(12):4142-4145.
[66]Zhao H,Malik H.Audio forensics using acoustic environment traces[C]//2012 IEEE Statistical Signal Processing Workshop,2012:373-376.
[67]钟巍,孔祥维,尤新刚,等.基于分数倒谱变换的取证语音拼接特征提取与分析[J].数据采集与处理,2014(2):248-253.
[68]Ikram S,Malik H.Digital audio forensics using background noise[C]//2010 IEEE International Conference on Multimedia and Expo(ICME),2010:106-110.
[69]阳帆,严迪群,徐宏伟,等.基于噪声一致性的数字语音异源拼接篡改检测算法[J].计算机应用,2017,37(12):3452-3457.
[70]Malik H.Acoustic environment identification and its applications to audio forensics[J].IEEE Transactions on Information Forensics and Security,2013,8(11):1827-1837.
[71]Zhao H,Malik H.Audio recording location identification using acoustic environment signature[J].IEEE Transactions on Information Forensics and Security,2013,8(11):1746-1759.
[72]Zhao H,Chen Y,Wang R,et al.Anti-forensics of environmental-signature-based audio splicing detection and its countermeasure via rich-features classification[J].IEEE Transactions on Information Forensics and Security,2016,11(7):1603-1617.
[73]Mascia M,Canclini A,Antonacci F,et al.Forensic and anti-forensic analysis of indoor/outdoor classifiers based on acoustic clues[C]//2015 23rd European Signal Processing Conference(EUSIPCO),2015:2072-2076.
[74]Farid H.Detecting digital forgeries using bispectral analysis[Z].Cambridge:MIT,1999.
[75]高阳,黄征,徐彻,等.基于高阶频谱分析的音频篡改鉴定[J].信息安全与通信保密,2008(2):94-96.
[76]Lin X,Kang X.Exposing speech tampering via spectral phase analysis[J].Digital Signal Processing,2017,60:63-74.
[77]林晓丹.基于广义高斯模型的音频篡改盲检测[J].计算机工程与设计,2013,34(9):3050-3062.
[78]林晓丹.采用线性预测模型的语音篡改检测[J].华侨大学大学报,2015,36(1):40-44.
[79]Chen J,Xiang S,Huang H,et al.Detecting and locating digital audio forgeries based on singularity analysis with wavelet packet[J].Multimedia Tools and Applications,2016,75(4):2303-2325.
[80]高华,包永强.基于时域和频域的语音篡改检测算法[J].声学技术,2014,33(4):119-122.
[81]潘平,罗辉,李换双.基于随机共振的语音真实性检测方法[J].计算机工程,2013,39(5):277-279.
[82]Zhou C,Pan P,Huang L.Authenticity identification of speaker digital recording data based on quantum genetic algorithm[J].Multimedia Tools&Applications,2017(4):1-15.
[83]Wang Z,Hong Q,Huang X.The construction and application of a speech database for voiceprint recognition[C]//15th Oriental COCOSDA Conference,Macau,China,2012.
[84]林晓丹,王佳斌.多特征联合的语音被动取证方法[J].微型机与应用,2012,31(20):39-41.
[85]Korycki R.Time and spectral analysis methods with machine learning for the authentication of digital audio recordings[J].Forensic Science International,2013,230(1):117-126.
[86]Luo D,Sun M,Huang J.Audio postprocessing detection based on amplitude cooccurrence vector feature[J].IEEESignal Processing Letters,2016,23(5):688-692.
[87]Fisher W M,Doddington G R,Goudie-Marshall K M.The DARPA speech recognition research database:specifications and status[C]//Proc DARPA Workshop on Speech Recognition,1986.
[88]Christopher C,Mark L.Issues in corpus creation and distribution:the evolution of the linguistic data consortium[J].Biochemical Pharmacology,2000.