数字图像来源取证现状与趋势

摘要/Abstract

摘要： 数字信息易编辑和伪造的特点，使得数字时代多媒体的真实性遭受了极大的挑战.正因为此，关注数字图像来源辨识性、内容真实性和信息完整性的数字图像取证技术成为多媒体信息内容安全研究的热点.针对数字图像的来源取证问题，首先阐述了已有的一般模型和框架，然后从设备类型、设备型号和设备个体取证鉴别3个层次，分别阐述了目前典型的研究思路和方法，进而在此基础上，分析和介绍了当前数字图像来源取证所面临的问题和发展趋势，最终给出了结论和展望.

关键词: 数字图像取证, 来源鉴别, 设备类型, 设备型号, 设备个体

Abstract: The characteristic of editing and forging digital media makes the authenticity of multi media encountering great challenges in digital age. Due to that, the digital image forensics, which focuses on the source identifiability, content authenticity and information integrity of digital images, has become a hot research issue in multimedia information security field. Focusing on the issue of source identification of digital images, the paper first illustrates the existing general research model and framework. Based on the overview of current typical algorithms from the perspective of device based, model based and camera based source identification respectively, the problems and trends of source identification of digital images are introduced and analyzed. Finally, this paper draws the conclusion and shows the prospect of digital image source identification.

Key words: digital image forensics, source identification, device based, model based, camera based

王波. 数字图像来源取证现状与趋势[J]. 信息安全研究, 2016, 2(6): 501-511.

参考文献

［1］Piva A. An overview on image forensics［J］. ISRN Signal Processing, 2013, 2013: 122［2］Stamm M C, Wu M, Liu K J R. Information forensics: An overview of the first decade［J］. IEEE Access, 2013, 1: 167200［3］Bestagini P, Fontani M, Milani S, et al. An overview of video forensics［J］. APSIPA Trans on Signal and Information Processing, 2012, 1(2): 12291233［4］Media Forensics(MediFor)［EBOL］. 2015［20160411］. https:www.fbo.govindex?s=opportunity&mode=form &id=bfa29e5f04566fbb961cd773a8a8649f&tab=core&_cview=1［5］Chang S F, Hsu J, Ng T T, et al. TrustFoto［EBOL］. University of Columbia, 2006［20160411］. http:www.ee.columbia.edulndvmmtrustfoto［6］Kot A C, Cao Hong. Image and video source class identification［M］. Digital Image Forensics. Berlin: Springer, 2013: 157178［7］孙晓婷, 李叶舟, 牛少彰, 等. 数字照片相机来源认证方法研究［J］. 中国电子商情通信市场, 2013 (6): 100104［8］王波. 利用成像引入特征的数字图像被动盲取证研究［D］. 大连: 大连理工大学, 2010［9］Lyu S, Farid H. How realistic is photorealistic?［J］. IEEE Trans on Signal Processing, 2005, 53(2): 845850［10］Ozparlak L, Avcibas I. Differentiating between images using waveletbased transforms: A comparative study［J］. IEEE Trans on Information Forensics and Security, 2011, 6(4): 14181431［11］Farid H, Bravo M J. Perceptual discrimination of computer generated and photographic faces［J］. Digital Investigation, 2012, 8(3／4): 226235［12］Ng T T, Chang S F, Hsu J, et al. Physicsmotivated features for distinguishing photographic images and computer graphics［C］ Proc of the 13th Annual ACM Int Conf on Multimedia. New York: ACM, 2005: 239248［13］Ng T T, Chang S F. An online system for classifying computer graphics images from natural photographs［C］ Proc of the Security, Steganography, and Watermarking of Multimedia Contents VIII. San Francisco, CA: SPIE, 2006: 6072116072119［14］Chen Wen, Shi Y Q, Xuan Guorong. Identifying computer graphics using HSV color model and statistical moments of characteristic functions［C］ Proc of the IEEE Int Conf on Multimedia and Expo. Piscataway, NJ: IEEE, 2007: 11231126［15］Chang T Y, Tai S C, Lin G S. A passive multipurpose scheme based on periodicity analysis of CFA artifacts for image forensics［J］. Journal of Visual Communication and Image Representation, 2014, 25(6): 12891298［16］Peng Fei, Liu Juan, Long Min. Identification of natural images and computer generated graphics based on hybrid features［J］. International Journal of Digital Crime and Forensics, 2012, 4(1): 116［17］Wang Xiaofeng, Liu Yong, Xu Bingchao, et al. A statistical feature based approach to distinguish PRCG from photographs［J］. Computer Vision and Image Understanding, 2014, 128(11): 8493［18］Peng Fei, Li Jiaoting, Long Min. Identification of natural images and computergenerated graphics based on statistical and textural features［J］. Journal of Forensic Sciences, 2015, 60(2): 435443［19］王学良, 李生红, 金波, 等. 一种用于计算机生成图像与自然图像鉴别的改进方法［J］. 光电子·激光, 2010, 21(5): 783785［20］刘娟. 基于PRNU的自然图像和计算机生成图像来源取证［D］. 长沙: 湖南大学, 2012［21］Gao Shang, Zhang Cong, Wu Chanle, et al. A hybrid feature based method for distinguishing computer graphics and photographic image［G］ LNCS 8389: Proc of the Int Workshop on DigitalForensics and Watermarking 2013. Berlin: Springer, 2013: 303313［22］Li Zhaohong, Zhang Zhenzhen, Shi Y Q. Distinguishing computer graphics from photographic images using a multiresolution approach based on local binary patterns［J］. Security and Communication Networks, 2014, 7(11): 21532159［23］Orozco A L S, Corripio J R, Villalba L J G, et al. Image source acquisition identification of mobile devices based on the use of features［J］. Multimedia Tools and Application, 2015: 125［24］Khanna N, Chiu G T C, Allebach J P, et al. Forensic Techniques for classifying scanner, computer generated and digital camera images［C］ Proc of the IEEE Int Conf on Acoustics, Speech and Processing. Piscataway, NJ: IEEE, 2008: 16531656［25］San Choi K, Lam E Y, Wong K K Y. Automatic source camera identification using the intrinsic lens radial distortion［J］. Optics Express, 2006, 14(24): 1155111565［26］Popescu A C, Farid H. Exposing digital forgeries in color filter array interpolated images［J］. IEEE Trans on Signal Processing, 2005, 53(10): 39483959［27］Bayram S, Sencarb H T, Memonb N. Classification of digital cameramodels based on demosaicing artifacts［J］. Digital Investigation, 2008, 5(12): 4959［28］Long Yangjing, Huang Yizhen. Image based source camera identification using demosaicking［C］ Proc of the IEEE Workshop on Multimedia Signal Processing. Piscataway, NJ: IEEE, 2006: 419424［29］Swaminathan A, Wu Min, Liu K J R. Nonintrusive component forensics of visual sensors using output images［J］. IEEE Trans on Information Forensics and security, 2007, 2(1): 91106［30］王波, 孔祥维, 尤新刚, 等. 利用协方差矩阵检测CFA插值的相机来源鉴别方法［J］. 光电子·激光, 2009, 20(4): 517520［31］Chen Chen, Stamm M C. Camera model identification framework using an ensemble of demosaicing features［C］ Proc of the Int Workshop on Information Forensics and Security. Piscataway, NJ: IEEE, 2015: 16［32］Deng Zhonghai, Gijsenij A, Zhang Jingyuan. Source camera identification using autowhite balance approximation［C］ Proc of the Int Conf on Computer Vision. Piscataway, NJ: IEEE, 2011: 5764［33］Kharrazi M, Sencar H T, Memon N. Blind source camera identification［C］ Proc of the Int Conf on Image Processing. Piscataway, NJ: IEEE, 2004: 709712［34］Goyal K, Panwar R, Khanna N. Evaluation of IQMs effectiveness for cell phone identification using captured videos and images［C］ Proc of the Int Conf on Power, Control and Embedded Systems. Piscataway, NJ: IEEE, 2014: 16［35］Meng Fanjie, Kong Xiangwei, You Xingang. A new featurebased method for source camera identification［C］ Advances in Digital Forensics IV: Proc of IFIP Int Federation for Information Processing. Berlin: Springer, 2008: 207218［36］Xu Guanshuo, Shi Yunqing, Su Wei. Camera brand and model identification using moments of 1D and 2D characteristic functions［C］ Proc of the 16th IEEE Int Conf on Image Processing. Piscataway, NJ: IEEE, 2009: 29172920［37］Razzazi F, Seyedadabi. A robust feature for single image camera identification using local binary patterns［C］ Proc of the IEEE Int Symp on Signal Processing and Information Technology. Piscataway, NJ: IEEE, 2014: 462467［38］Thai T H, Cogranne R, Retraint F. Camera model identification based on the heteroscedastic noise model［J］. IEEE Trans on Image Processing, 2014, 23(1): 250263［39］Amerini I, Caldelli R, Crescenzi P, et al. Blind image clustering based on the normalized cuts critertion for camera identification［J］. Signal Processing: Image Communication, 2014, 29(8): 831843［40］Luan Shuhan, Kong Xiangwei, Wang Bo, et al. Silhouette coefficient based approach on cellphone classification for unknown source images［C］ Proc of the 2012 IEEE Int Conf on Communications. Piscataway, NJ: IEEE, 2012: 67446747［41］Geradts Z J, Bijhold J, Kief M, et al. Methods for identification of images acquired with digital cameras［C］ Proc of the SPIE Conf on Enabling Technologies for Law Enforcement and Security. Boston, MA, USA: SPIE, 2001: 505512［42］Dirik A E, Sencar H T, Memon N. Source camera identification based on sensor dust characteristics［C］ Proc of the IEEE Workshop on Signal Processing Applications for Public Security and Forensics. Piscataway, NJ: IEEE, 2007: 16［43］Fridrich J. Sensor Defects in Digital Image Forensic［M］. Berlin: Springer, 2013: 179218［44］Fridrich J. Digital image forensics using sensor noise［J］. IEEE Signal Processing Magazine, 2009, 26(2): 2637［45］Lawgaly A, Khelifi F, Bouridane. A Weighted averagingbased sensor pattern noise estimation for source camera identification［C］ Proc of the IEEE Int Conf on Image Processing. Piscataway, NJ: IEEE, 2014: 53575361［46］Tan Yue, Wang Bo, Zhao Meijuan, et al. Patchbased sensor pattern noise for camera source identification［C］ Proc of the IEEE China Summit and Int Signal and Information Processing. Piscataway, NJ: IEEE, 2015: 866870［47］Kang Xiangui, Chen Jiansheng, Lin Kerui, et al. A contextadaptive SPN predictor for trustworthy source camera identification［J］. EURASIP Journal on Image and Video Processing, 2014, 2014(1): 111［48］Hu Yongjian, Yu Binghua, Jian Chao. Source camera identification using large components of sensor pattern noise［C］ Proc of the 2nd Int Conf on Computer Science and its Applications. Piscataway, NJ: IEEE, 2009: 15［49］Li C T, Li Yue. Digital camera identification using colourdecoupled photo response nonuniformity noise pattern［C］ Proc of IEEE Int Symp on Circuits and Systems. Piscataway, NJ: IEEE, 2010: 30523055［50］Goljan M, Fridrich J, Chen Mo. Defending against fingerprintcopy attack in sensorbased camera identification［J］. IEEE Trans on Information Forensics and Security, 2011, 6(1): 227236［51］Lin Xufeng, Li C T. Preprocessing reference sensor pattern noise via spectrum equalization［J］. IEEE Trans on Information Forensics and Security, 2016, 11(1): 126140［52］Lin Xufeng, Li C T. Enhancing sensor pattern noise via filtering distortion removal［J］. IEEE Signal Processing Letters, 2016, 23(3): 381385［53］Chan L H, Law N F, Siu W C. A confidence map and pixelbased weighted correlation for PRNUbased camera identification［J］. Digital Investigation, 2013, 10(3): 215225［54］Li Ruizhe, Yu Guan, Li C T. PCAbased denoising of sensor pattern noise for source camera identification［C］ Proc of the IEEE China Summit and Int Signal and Information Processing. Piscataway, NJ: IEEE, 2014: 436440［55］Kang Xiangui, Li Yinxiang, Qu Zhenhua, et al. Enhancing source camera identification performance with a camera reference phase sensor pattern noise［J］. IEEE Trans on Information Forensics and Security, 2012, 7(2): 393402［56］Soobhany, Lam K P, Fletcher P, et al. Mobile camera source identification with SVD［G］LNEE 313: Proc of the Int Joint Conf on Computer Information and Systems Sciences and Engineering. Berlin: Springer, 2015: 123131［57］Chen Mo, Fridrich J, Goljan M, et al. Source digital camcorder identification using sensor photo response nonuniformity［C］ Proc of the 9th Conf on Security, Steganography, and Watermarking of Multimedia Contents IX. San Francisco, CA: SPIE, 2007: 65051G65051G12［58］Gou Hongmei, Swaminathan A, Wu Min. Intrinsic sensor noise features for forensic analysis on scanners and scanned images［J］. IEEE Trans on Information Forensics and Security, 2009, 4(3): 476491［59］Huang Yizhen. Can digital image forgery detection be unevadable? A case study: Color filter array interpolation statistical feature recovery［C］ Proc of the Visual Communications and Image Processing. San Francisco, CA: SPIE, 2005: 59602W59602W12［60］Kirchner M, Bohme R. Synthesis of color filter array pattern in digital images［C］ Proc of the Media Forensics and Security. San Francisco, CA: SPIE, 2009: 72540K72540K14［61］Li C T, Chang C Y, Li Yue. On the Repudiability of Device Identification and Image Integrity Verification Using Sensor Pattern Noise［G］ LNCS 45: Proc of the Int Conf on Information Security and Digital Forensics. Berlin: Springer, 2010: 1529［62］杨弘，周治平. 数字图像模式噪声篡改反取证［J］. 计算机工程与应用, 2014, 50(18): 156161［63］Zeng Hui, Jiang Yunwen, Kang Xiangui, et al. Game theoretic analysis of camera source identification［C］ Proc of the AsiaPacific Signal and Information Processing Association Annual Summit and Conf. Piscataway, NJ: IEEE, 2013: 19［64］Zeng Hui, Chen Jiansheng, Kang Xiangui, et al. Removing camera fingerprint to disguise photograph source［C］ Proc of the IEEE Int Conf on Image Processing. Piscataway, NJ: IEEE, 2015: 16871691［65］Caldelli R, Amerini I, Novi A. An analysis on attacker actions in fingerprintcopy attack in source camera identification［C］ Proc of the IEEE Int Workshop on Information Forensics and Security. Piscataway, NJ: IEEE, 2011: 16［66］Rao Quanquan, Luo Weiqi, Li Haodong, et al. Antiforensics of the triangle test by random fingerprintcopy attack［C］ Proc of Computational Visual Media Conf. Piscataway, NJ: IEEE, 2013: 16［67］Marra F, Roli F, Cozzolino D, et al. Attacking the triangle test in sensorbased camera identification［C］ Proc of the IEEE Int Conf on Image Processing. Piscataway, NJ: IEEE, 2014: 53075311［68］Barni M. A game theoretic approach to source identification with known statistics［C］ Proc of the IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2012: 17451748［69］Stamm M C, Lin W S, Liu K J R. Forensics vs. antiforensics: A decision and game theoretic framework［C］ Proc of the IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2012: 17491752［70］Wang Bo, Tan Yue, Zhao Meijuan, et al. Classifier combination based source identification for cell phone images［J］. KSII Trans on Internet and Information Systems, 2015, 9(12): 50875102［71］Costa F d O, Silva E, Eckmann M, et al. Open set source camera attribution and device linking［J］. Pattern Recognition Letters, 2014, 39(1): 92101［72］Huang Yonggang, Zhang Jun, Huang Heyan. Camera model identification with unknown models［J］. IEEE Trans on Information Forensics and Security, 2015, 10(12): 26922704［73］王波, 孔祥维, 尤新刚, 等. 基于协方差矩阵的CFA插值盲检测方法［J］. 电子与信息学报, 2009, 31(5): 11751179［74］Miroslav Goljan, Mo Chen, Pedro Comesana, et al. Effect of compression on sensorfingerprint based camera identification［C］ Proc of the Media Watermarking, Security, and Forensics. San Francisco, CA: SPIE, 2016［75］Goljan M, Fridrich J. Camera identification from cropped and scaled images［C］ Proc of the Security, Forensics, Steganography, and Watermarking of Multimedia Contents X. San Francisco, CA: SPIE, 2008: 68190E［76］Goljan M, Fridrich J. Sensor fingerprint digests for fast camera identification from geometrically distorted images［C］ Proc of the Media Watermarking, Security, and Forensics 2013. San Francisco, CA: SPIE, 2013: 866508［77］Goljan M, Fridrich J, Luká J. Camera identification from printed images［C］ Proc of the Security, Forensics, Steganography, and Watermarking of Multimedia Contents X. San Francisco, CA: SPIE, 2008: 68190I68190I12［78］Goljan M, Fridrich J, Filler T. Large scale test of sensor fingerprint camera identification［C］ Proc of the Media Forensics and Security XI. San Francisco, CA: SPIE, 2009: 72540172540112［79］谭跃, 王波, 赵美娟. 有限样本条件下的相机来源鉴别方法［C］ 第十二届全国信息隐藏暨多媒体信息安全学术大会论文集, 2015: 448454［80］Tan Yue, Wang Bo, Li Ming, et al. Camera source identification with limited labeled training set［G］ LNCS 9569: Proc of the 14th Int Workshop. Berlin: Springer, 2016: 1827