基于“鬼影”的数字视频篡改检测
|
摘要
随着复杂低耗数字视频摄像机的广泛使用,以及视频共享网站的普及,数字视频在我们的日常生活中扮演的角色越来越重要。但各种视频造假事件,也严重降低了视频的公共置信度。因此,检验数字视频的完整性和原始性具有重要的学术意义和实用价值。
本论文针对数字视频盲取证技术进行深入研究,数字视频盲取证技术是根据视频数据自身来判断视频是否被篡改的取证技术,本文综述了数字视频盲取证技术的研究现状,介绍了数字视频的篡改方式及相应的篡改工具,在此基础上提出了一种新颖的基于视频修复痕迹——“鬼影”的视频篡改检测算法。对于移除视频中目标运动物体的篡改,由于移除物体后需采用修复技术对移除目标区域进行修补,而视频修复往往很难确保运动轨迹的连续性和一致性,使得篡改后的视频遗留有修复痕迹——“鬼影”。本文算法首先由累积帧差得到运动前景的运动轨迹,而后用块匹配法建立运动前景的拼接图,如果运动轨迹和前景拼接图不一致,待测视频就被判定为带有修复痕迹——“鬼影”的篡改视频,否则,则认为视频未被篡改。实验结果表明算法能够较好地检测出视频的篡改,且对有损压缩具有鲁棒性。
本论文所解决的是一个具有挑战性的问题,在理论和应用领域都具有重大意义。本论文主要采取数据分析、理论推演以及计算机仿真的研究方法,主要的仿真工具有matlab 7.6.0, VC++等。
With the wide-spread availability of sophisticated and low-cost digital video cameras and the prevalence of video sharing websites, digital videos are playing a more important role in our daily life. But the fact that all kinds of videos are easily tampered has seriously reduced the public confidence in videos. So it is importmant to ensure the integrity and authenticity of the digital videos both for important academic significance and practical values.
The thesis focuses and studies on digital video forensics, which is a new technology used to judge the videos whether they are inpainted or not according to the video data. The present situation of digital video forensics is discussed and tampered methods and tools are introduced, and then a novel approach is proposed for detecting video forgery based on exposing ghost shadow artifact. When tampering video by removing the moving object, it is necessary to complete the forged region by video inpainting. But video inpainting usually cannot ensure the consistence and continuity of the motion trace. Thus the ghost shadow artifact is introduced in the forged video. In our approach, the track of moving foreground is obtained by accumulative differences and the moving foreground mosaic is built by block matching. If the moving track and the foreground mosaic are inconsistent, the input video is judged as forged video with ghost shadow artifact. Otherwise, it is judged as authentic video. Experiments show that our approach can accurately detect the forgery in the video, and it's robust to the lossy compression.
Digital video forensics is a challenging question, and it's significant in theories and applications. In this thesis, data analysis, theory deduction and computer simulation methods are utilized for studying. The simulation tools include matlab 7.6.0, VC++, and so on.
本论文针对数字视频盲取证技术进行深入研究,数字视频盲取证技术是根据视频数据自身来判断视频是否被篡改的取证技术,本文综述了数字视频盲取证技术的研究现状,介绍了数字视频的篡改方式及相应的篡改工具,在此基础上提出了一种新颖的基于视频修复痕迹——“鬼影”的视频篡改检测算法。对于移除视频中目标运动物体的篡改,由于移除物体后需采用修复技术对移除目标区域进行修补,而视频修复往往很难确保运动轨迹的连续性和一致性,使得篡改后的视频遗留有修复痕迹——“鬼影”。本文算法首先由累积帧差得到运动前景的运动轨迹,而后用块匹配法建立运动前景的拼接图,如果运动轨迹和前景拼接图不一致,待测视频就被判定为带有修复痕迹——“鬼影”的篡改视频,否则,则认为视频未被篡改。实验结果表明算法能够较好地检测出视频的篡改,且对有损压缩具有鲁棒性。
本论文所解决的是一个具有挑战性的问题,在理论和应用领域都具有重大意义。本论文主要采取数据分析、理论推演以及计算机仿真的研究方法,主要的仿真工具有matlab 7.6.0, VC++等。
With the wide-spread availability of sophisticated and low-cost digital video cameras and the prevalence of video sharing websites, digital videos are playing a more important role in our daily life. But the fact that all kinds of videos are easily tampered has seriously reduced the public confidence in videos. So it is importmant to ensure the integrity and authenticity of the digital videos both for important academic significance and practical values.
The thesis focuses and studies on digital video forensics, which is a new technology used to judge the videos whether they are inpainted or not according to the video data. The present situation of digital video forensics is discussed and tampered methods and tools are introduced, and then a novel approach is proposed for detecting video forgery based on exposing ghost shadow artifact. When tampering video by removing the moving object, it is necessary to complete the forged region by video inpainting. But video inpainting usually cannot ensure the consistence and continuity of the motion trace. Thus the ghost shadow artifact is introduced in the forged video. In our approach, the track of moving foreground is obtained by accumulative differences and the moving foreground mosaic is built by block matching. If the moving track and the foreground mosaic are inconsistent, the input video is judged as forged video with ghost shadow artifact. Otherwise, it is judged as authentic video. Experiments show that our approach can accurately detect the forgery in the video, and it's robust to the lossy compression.
Digital video forensics is a challenging question, and it's significant in theories and applications. In this thesis, data analysis, theory deduction and computer simulation methods are utilized for studying. The simulation tools include matlab 7.6.0, VC++, and so on.
引文
[1]吴金海,林福宗,基于数字水印的图像认证技术,计算机学报[J],2004,27(9)：1153-1161.
[2]Ng T T, Chang S F, Lin C Y, Qibin Sun Q B, Passive-blind image forensics, in: Multimedia Security Technologies for Digital Rights[M], New York:Elsevier,2006.
[3]Tran V L, Chong, K S, Emmanuel S, Kankanhalli M S, A survey on Digital Camera Image Forensic Methods [C], in:Proceedings of IEEE International Conference on Multimedia and Expo,2007.07,16-19.
[4]Choi K S, Lam E Y, Wong K K Y, Source camera identification using footprints from lens aberration, Proceedings of the SPIE,2006.
[5]Choi K S, Lam E Y, Wong K K Y, Automatic source camera identification using the intrinsic lens radial distortion, Optics Express,2006,14(24):11551-11565.
[6]Geradts ZJ, Bijhold J, Kieft M, Kurosawa K, Kuroki K, Saitoh N, Methods for identification of images acquired with digital cameras [C], Processing of SPIE, Enabling Technologies for Law Enforcement and Security,2000,4232:505-512.
[7]Lukas J, Fridrich J, Goljan M, Digital camera identification from sensor pattern noise [J], IEEE Transactions on Information Forensics and Security,2006.06,1(2): 205-214.
[8]Bayram S, Sencar H, Memon N, Avcibas, Source camera identification based on CFA interpolation, in:Proceedings of IEEE International Conference on Image Processing [C],2005,12:69-72.
[9]Ng T T, Chang S F, Tsui M P, Using geometry invariants for camera response function estimation, IEEE Computer Society Conference on Computer Vision and Pattern Recognition [C],2007.06,1-8.
[10]Mehdi K L, Sencar H T, Memon N, Blind source camera identification, International Conference on Image Processing [C],2004.10,1:709-712.
[11]Tsai M J, Wu G H, Using image features to identify camera sources, ICASSP, 2006.
[12]Farid H, Digital image ballistics from JPEG quantization, Technical Report, TR2006-583, Dartmouth College, Computer Science.
[13]Choi K S, Lam E Y, Wong, Kenneth K.Y, Source Camera Identification by JPEG Compression Statistics for Image Forensics, TENCON, IEEE Region 10 Conference [C],2006.11,1-4.
[14]吴琼,李国辉,涂丹,等,面向真实性鉴别的数字图像盲取证技术综述[J],自动化学报,2008,34(12).
[15]Popescu A C, Farid H, Exposing digital forgeries by detecting traces of resampling [J], IEEE Transactions on Signal Processing,2005,53(2):758-767.
[16]Fridrich J, Lukas J, Estimation of primary quantization matrix in double compressed JPEG images, in:Proceedings of Digital Forensic Research Workshop, Cleveland, OH, USA,2003.
[17]Adams J, Parulski K, Spaulding K, Color processing in digital cameras, IEEE Micro,1998,18(6):20-30.
[18]Lyu S W, Farid H, How realistic is photorealistic [J], IEEE Transactions on Signal Processing,2005,53(2):845-850.
[19]Lyu S W, Natural Image Statistics for Digital Image Forensics [Ph.D. dissertation], Department of Computer Science, Darthmouth College,2005.
[20]Popescu A C, Farid H, Statistical Tools for Digital Forensics,6th International Workshop on Information Hiding, Toronto, Canada,2004,128-147.
[21]Wang W, Farid H, Exposing digital forgeries in video by detecting double MPEG Compression, in:Proceedings of the Multimedia and Security Workshop [C], Geneva, Switzerland,2006,37-47.
[22]Wang W, Farid H, Exposing digital forgeries in video by detecting duplication, in:Proceedings of the Multimedia and Security Workshop [C], Dallas, TX,2007, 35-42.
[23]Wang W, Farid H, Exposing digital forgeries in interlaced and de-Interlaced Video [J], IEEE Transactions on Information Forensics and Security,2007,2(3): 438-449.
[24]Kobayashi M, Okabe T, Sato Y, Detecting video forgeries based on noise characteristics, in:Proceedings of the 3rd Pacific-Rim Symposium on Image and Video Technology [J], Tokyo, Japan,2009, LNCS 5414,306-317.
[25]De A, Chadha H, Gupta S, Detection of forgery in digital video, in:Proceedings of the 10th World Multi-Conference on Systemics, Cybernetics and Informatics [C], 2006,5:229-233.
[26]Hsu C, Hung T, Lin C, Hsu C, Video forgery detection using correlation of noise residue, in:Proceedings of IEEE Workshop Multimedia Signal Processing (MMSP) [C], Cairns, Queensland, Australia,2008.10,170-174.
[27]张春田,苏育挺,张静,等,数字图像压缩编码[M],北京：清华大学出版社,2006.01,298-316.
[28]周欣,MPEG标准和MPEG-2视频编码系统[J],无线技术,2009.07.
[29]钟玉琢,王琪,赵黎,MPEG-2运动图像压缩编码国际标准及MPEG的新进展[M],北京：清华大学出版社,2002.03,102-110.
[30]张红英,彭启宗,数字图像修复技术综述[J],中国图象图形学报,2007,12(1)：1-10.
[31]吴琼,孙韶杰,朱为,等,数字图像盲取证对样本合成修复应用的篡改区域检测算法[J],自动化学报,2007.12,35(3)：239-243.
[32]姚建亮,彭宏京,一种改进的基于样本的图像修复算法[J],电子科技,2010,01,100-103.
[33]刘旭,王慈,等,自适应多方向视频图像修复技术[J],微计算机信息,2009,28,112-114.
[34]Wexler Y, Shechtman E, Irani M, Space-time completion of video [J], IEEE Trans. Pattern Anal. Mach. Intell,2007,29(3):463-476.
[35]Patwardhan K, Sapiro A G, Bertalmio M, Video inpainting under constrained camera motion [J], IEEE Trans on Image Process,2007,16(2):545-553.
[36]朱为,李国辉,涂丹,等,基于纹理合成的视频去抖动[J],计算机工程与应用,2008,44(32).
[37]Shih T K, Tang N C, Yeh W S, Chen T J, Lee W, Video inpainting and implant via diversified temporal continuations [C], in Proc.ACM Int. Conf. Multimedia, 2006.10,133-136.
[38]Shih T K, Tang N C, Hwang J N, Ghost shadow removal in multi-layered video inpainting [C], in Proc. IEEE Int. Conf. Multimedia& Expo,2007.07,1471-1474.
[39]Shih T K, Tang N C, Hwang J N, Exemplar-based video inpainting without ghost shadow artifact by maintaining temporal continuity [J], IEEE Transactions on Circuits and Systems for Video Technology,2009,19(3):347-360.
[40]Gonzalez R C, Woods R E, Digital Image Processing, Second Edition, Beijing: Publishing House of Electron-ics Industry,2002,523-556.
[2]Ng T T, Chang S F, Lin C Y, Qibin Sun Q B, Passive-blind image forensics, in: Multimedia Security Technologies for Digital Rights[M], New York:Elsevier,2006.
[3]Tran V L, Chong, K S, Emmanuel S, Kankanhalli M S, A survey on Digital Camera Image Forensic Methods [C], in:Proceedings of IEEE International Conference on Multimedia and Expo,2007.07,16-19.
[4]Choi K S, Lam E Y, Wong K K Y, Source camera identification using footprints from lens aberration, Proceedings of the SPIE,2006.
[5]Choi K S, Lam E Y, Wong K K Y, Automatic source camera identification using the intrinsic lens radial distortion, Optics Express,2006,14(24):11551-11565.
[6]Geradts ZJ, Bijhold J, Kieft M, Kurosawa K, Kuroki K, Saitoh N, Methods for identification of images acquired with digital cameras [C], Processing of SPIE, Enabling Technologies for Law Enforcement and Security,2000,4232:505-512.
[7]Lukas J, Fridrich J, Goljan M, Digital camera identification from sensor pattern noise [J], IEEE Transactions on Information Forensics and Security,2006.06,1(2): 205-214.
[8]Bayram S, Sencar H, Memon N, Avcibas, Source camera identification based on CFA interpolation, in:Proceedings of IEEE International Conference on Image Processing [C],2005,12:69-72.
[9]Ng T T, Chang S F, Tsui M P, Using geometry invariants for camera response function estimation, IEEE Computer Society Conference on Computer Vision and Pattern Recognition [C],2007.06,1-8.
[10]Mehdi K L, Sencar H T, Memon N, Blind source camera identification, International Conference on Image Processing [C],2004.10,1:709-712.
[11]Tsai M J, Wu G H, Using image features to identify camera sources, ICASSP, 2006.
[12]Farid H, Digital image ballistics from JPEG quantization, Technical Report, TR2006-583, Dartmouth College, Computer Science.
[13]Choi K S, Lam E Y, Wong, Kenneth K.Y, Source Camera Identification by JPEG Compression Statistics for Image Forensics, TENCON, IEEE Region 10 Conference [C],2006.11,1-4.
[14]吴琼,李国辉,涂丹,等,面向真实性鉴别的数字图像盲取证技术综述[J],自动化学报,2008,34(12).
[15]Popescu A C, Farid H, Exposing digital forgeries by detecting traces of resampling [J], IEEE Transactions on Signal Processing,2005,53(2):758-767.
[16]Fridrich J, Lukas J, Estimation of primary quantization matrix in double compressed JPEG images, in:Proceedings of Digital Forensic Research Workshop, Cleveland, OH, USA,2003.
[17]Adams J, Parulski K, Spaulding K, Color processing in digital cameras, IEEE Micro,1998,18(6):20-30.
[18]Lyu S W, Farid H, How realistic is photorealistic [J], IEEE Transactions on Signal Processing,2005,53(2):845-850.
[19]Lyu S W, Natural Image Statistics for Digital Image Forensics [Ph.D. dissertation], Department of Computer Science, Darthmouth College,2005.
[20]Popescu A C, Farid H, Statistical Tools for Digital Forensics,6th International Workshop on Information Hiding, Toronto, Canada,2004,128-147.
[21]Wang W, Farid H, Exposing digital forgeries in video by detecting double MPEG Compression, in:Proceedings of the Multimedia and Security Workshop [C], Geneva, Switzerland,2006,37-47.
[22]Wang W, Farid H, Exposing digital forgeries in video by detecting duplication, in:Proceedings of the Multimedia and Security Workshop [C], Dallas, TX,2007, 35-42.
[23]Wang W, Farid H, Exposing digital forgeries in interlaced and de-Interlaced Video [J], IEEE Transactions on Information Forensics and Security,2007,2(3): 438-449.
[24]Kobayashi M, Okabe T, Sato Y, Detecting video forgeries based on noise characteristics, in:Proceedings of the 3rd Pacific-Rim Symposium on Image and Video Technology [J], Tokyo, Japan,2009, LNCS 5414,306-317.
[25]De A, Chadha H, Gupta S, Detection of forgery in digital video, in:Proceedings of the 10th World Multi-Conference on Systemics, Cybernetics and Informatics [C], 2006,5:229-233.
[26]Hsu C, Hung T, Lin C, Hsu C, Video forgery detection using correlation of noise residue, in:Proceedings of IEEE Workshop Multimedia Signal Processing (MMSP) [C], Cairns, Queensland, Australia,2008.10,170-174.
[27]张春田,苏育挺,张静,等,数字图像压缩编码[M],北京：清华大学出版社,2006.01,298-316.
[28]周欣,MPEG标准和MPEG-2视频编码系统[J],无线技术,2009.07.
[29]钟玉琢,王琪,赵黎,MPEG-2运动图像压缩编码国际标准及MPEG的新进展[M],北京：清华大学出版社,2002.03,102-110.
[30]张红英,彭启宗,数字图像修复技术综述[J],中国图象图形学报,2007,12(1)：1-10.
[31]吴琼,孙韶杰,朱为,等,数字图像盲取证对样本合成修复应用的篡改区域检测算法[J],自动化学报,2007.12,35(3)：239-243.
[32]姚建亮,彭宏京,一种改进的基于样本的图像修复算法[J],电子科技,2010,01,100-103.
[33]刘旭,王慈,等,自适应多方向视频图像修复技术[J],微计算机信息,2009,28,112-114.
[34]Wexler Y, Shechtman E, Irani M, Space-time completion of video [J], IEEE Trans. Pattern Anal. Mach. Intell,2007,29(3):463-476.
[35]Patwardhan K, Sapiro A G, Bertalmio M, Video inpainting under constrained camera motion [J], IEEE Trans on Image Process,2007,16(2):545-553.
[36]朱为,李国辉,涂丹,等,基于纹理合成的视频去抖动[J],计算机工程与应用,2008,44(32).
[37]Shih T K, Tang N C, Yeh W S, Chen T J, Lee W, Video inpainting and implant via diversified temporal continuations [C], in Proc.ACM Int. Conf. Multimedia, 2006.10,133-136.
[38]Shih T K, Tang N C, Hwang J N, Ghost shadow removal in multi-layered video inpainting [C], in Proc. IEEE Int. Conf. Multimedia& Expo,2007.07,1471-1474.
[39]Shih T K, Tang N C, Hwang J N, Exemplar-based video inpainting without ghost shadow artifact by maintaining temporal continuity [J], IEEE Transactions on Circuits and Systems for Video Technology,2009,19(3):347-360.
[40]Gonzalez R C, Woods R E, Digital Image Processing, Second Edition, Beijing: Publishing House of Electron-ics Industry,2002,523-556.