计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (6): 187-199.doi: 10.11896/jsjkx.241100190

• 计算机图形学&多媒体 • 上一篇    下一篇

面向开放环境的PRNU指纹提纯算法

刘宇飞, 肖延辉, 田华伟   

  1. 中国人民公安大学国家安全学院 北京 100038
  • 收稿日期:2024-11-29 修回日期:2025-02-16 出版日期:2025-06-15 发布日期:2025-06-11
  • 通讯作者: 肖延辉(xiaoyanhui@ppsuc.edu.cn)
  • 作者简介:(2022211246@stu.ppsuc.edu.cn)
  • 基金资助:
    公安部科技计划技术研究项目(2022JSYJC22)

PRNU Fingerprint Purification Algorithm for Open Environment

LIU Yufei, XIAO Yanhui, TIAN Huawei   

  1. School of National Security,People's Public Security University of China,Beijing 100038,China
  • Received:2024-11-29 Revised:2025-02-16 Online:2025-06-15 Published:2025-06-11
  • About author:LIU Yufei,born in 1999,postgraduate,is a member of CCF(No.U9221G).His main research interest include multimedia forensics.
    XIAO Yanhui,born in 1984,Ph.D,associated professor,Ph.D supervisor.His main research interest is multimedia forensics.
  • Supported by:
    Ministry of Public Security Science and Technology Plan Technical Research Project(2022JSYJC22).

PDF (PC)

109

摘要： 数字图像后处理流程带来的非唯一性人造(Non-Unique Artifacts,NUAs)噪声掺杂在具有唯一性、稳定性的光响应非均质性(Photo-Response Non-Uniformity,PRNU)指纹中,极大地影响了下游成像设备溯源任务的精确性。然而,现有NUAs抑制方案主要针对实验环境,不仅需要额外的超参数设定,而且需额外的算力和存储空间,难以在开放环境中实际应用。为解决该问题,提出了一种面向开放环境的PRNU指纹提纯算法。首先,对现有PRNU指纹相关性度量指标即峰值相关能量比(Peak-to-Correlation Energy Ratio,PCE)进行改进,提出了基于归一化的PCE_norm和PCE_denuas,以实现开放环境下的自适应相关性度量。然后,通过构建对比学习机制缩小同一指纹和放大不同指纹的距离,实现NUAs离线抑制,从而在溯源任务中不需额外计算和存储成本进行在线抑制。最后,通过在Dresden和Daxing数据集上的实验证明了所提算法的有效性和鲁棒性。

关键词: 成像设备溯源, PRNU指纹, NUAs噪声, 相关性度量, 对比学习, 提纯算法

Abstract: Non-unique artifacts(NUAs) noise generated by digital image post-processing pipeline is mixed in unique and stable photo response non-uniformity(PRNU) fingerprints.It seriously affects the precision of the downstream source camera identification(SCI) task.However,existing NUAs suppression schemes mainly target experimental environments and require not only additional hyperparameter settings,but also additional computing resources and storage space,which are difficult to apply in open environments.To solve this problem,this paper proposes a PRNU fingerprint purification algorithm for open environments.Firstly,it improves the existing PRNU fingerprint correlation metric peak-to-correlation energy ratio(PCE) and proposes PCE_norm and PCE_denuas based on normalization to achieve adaptive correlation measurement in open environment.Then,NUAs offline suppression is realized by constructing a contrastive learning mechanism to reduce the distance of the same fingerprint and amplify different fingerprints,so that there is no need for additional computation and storage costs for online suppression in SCI tasks.Finally,experiments on Dresden and Daxing datasets demonstrate the effectiveness and robustness of the proposed algorithm.

Key words: Source camera identification, PRNU fingerprint, NUAs noise, Correlation metric, Contrastive learning, Purification algorithm

中图分类号: 

  • TP391
[1]GAO C F,XIAO Y H,TIAN H W.Image camera fingerprint extraction algorithm based on MPRNet[J].Netinfo Security,2022,22(10):15-23.
[2]MANI R G,PARTHASARATHY R,ESWARAN S,et al.Asurvey on digital image forensics:Metadata and image forgeries[C]//Workshop on Applied Computing.2022:22-55.
[3]LUKAS J,FRIDRICH J,GOLJAN M.Digital camera identification from sensor pattern noise[J].IEEE Transactions on Information Forensics and Security,2006,1(2):205-214.
[4]JIANG X,WEI S K,ZHAO R Z,et al.Survey of imaging device source identification[J].Journal of Beijing Jiaotong University,2019,43(2):48-57.
[5]VALSESIA D,COLUCCIA G,BIANCHI T,et al.User Authentication via PRNU-Based Physical Unclonable Functions[J].IEEE Transactions on Information Forensics and Security,2017,12(8):1941-1956.
[6]BA Z,PIAO S,FU X,et al.ABC:Enabling Smartphone Authentication with Built-in Camera[C]//Proceedings of the Network and Distributed System Security Symposium.2018.
[7]MOHANTY M,ZHANG M,ASGHAR M R,et al.e-PRNU:Encrypted Domain PRNU-Based Camera Attribution for Preserving Privacy[J].IEEE Transactions on Dependable and Secure Computing,2021,18(1):426-437.
[8]KORUS P,MEMON N.Computational Sensor Fingerprints[J].IEEE Transactions on Information Forensics and Security,2022,17:2508-2523.
[9]LIU L,FU X,CHEN X,et al.FITS:Matching camera fingerprints subject to software noise pollution[C]//Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security.New York:ACM,2023:1660-1674.
[10]MONTIBELLER A,PÉREZ-GONZÁLEZ F.An adaptive methodfor camera attribution under complex radial distortion corrections[J].IEEE Transactions on Information Forensics and Security,2023,19:385-400.
[11]GUPTA B,TIWARI M.Improving performance of source-ca-mera identification by suppressing peaks and eliminating low-frequency defects of reference SPN[J].IEEE Signal Processing Letters,2018,25(9):1340-1343.
[12]CHEN M,FRIDRICH J,GOLJAN M,et al.Determining image origin and integrity using sensor noise[J].IEEE Transactions on Information Forensics and Security,2008,3(1):74-90.
[13]LIN X,LI C T.Preprocessing reference sensor pattern noise via spectrum equalization[J].IEEE Transactions on Information Forensics and Security,2015,11(1):126-140.
[14]RAO Q,WANG J.Suppressing random artifacts in referencesensor pattern noise via decorrelation[J].IEEE Signal Proces-sing Letters,2017,24(6):809-813.
[15]ALBISANI C,IULIANI M,PIVA A.Checking PRNU Usability on modern devices[C]//ICASSP 2021-2021 IEEE International Conference on Acoustics,Speech and Signal Processing.IEEE,2021:2535-2539
[16]BHAT N N,BIANCHI T.Investigating inconsistencies in prnu-based camera identification[C]//Proceedings of 2022 IEEE International Conference on Image Processing.IEEE,2022:851-855.
[17]GOLJAN M.Digital camera identification from images-estimating false acceptance probability[C]//International workshop on digital watermarking.Berlin:Springer,2008:454-468.
[18]KANG X,LI Y,QU Z,et al.Enhancing source camera identification performance with a camera reference phase sensor pattern noise[J].IEEE Transactions on Information Forensics and Security,2011,7(2):393-402.
[19]LE-KHAC P H,HEALY G,SMEATON A F.Contrastive representation learning:A framework and review[J].IEEE Access,2020,8:193907-193934.
[20]HE K,FAN H,WU Y,et al.Momentum contrast for unsupervised visual representation learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.IEEE,2020:9726-9735.
[21]GAO T,YAO X,CHEN D.Simcse:Simple contrastive learning of sentence embeddings[J].arXiv:2104.08821,2021.
[22]SHEN D,ZHAO Y,QUAN C.Identity-Referenced DeepfakeDetection with Contrastive Learning[C]//Proceedings of the 2022 ACM Workshop on Information Hiding and Multimedia Security.New York:ACM,2022:27-32.
[23]ZHENG J,ZHOU Y,HU X,et al.Deepfake Detection withCombined Unsupervised-Supervised Contrastive Learning[C]//2024 IEEE International Conference on Image Processing.IEEE,2024:787-793.
[24]QIAO T,XIE S,CHEN Y,et al.Fully Unsupervised Deepfake Video Detection Via Enhanced Contrastive Learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2024,46(7):4654-4668.
[25]LIU X,YU Y,LI X,et al.MCL:Multimodal Contrastive Learning for Deepfake Detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2024,34(4):2803-2813.
[26]ZHU Y,ZHANG C,GAO J,et al.High-compressed deepfake video detection with contrastive spatiotemporal distillation[J].Neurocomputing,2024,565:126872.
[27]COZZOLINO D,VERDOLIVA L.Noiseprint:A CNN-BasedCamera Model Fingerprint[J].IEEE Transactions on Information Forensics and Security,2020,15:144-159.
[28]KALANTIDIS Y,SARIYILDIZ M B,PION N,et al.Hard negative mixing for contrastive learning[J].Advances in Neural Information Processing Systems,2020,33:21798-21809.
[29]TIAN Y,KRISHNAN D,ISOLA P.Contrastive multiview co-ding[C]//European Conference on Computer Vision.Cham:Springer,2020:776-794.
[30]CHEN X,FAN H,GIRSHICK R,et al.Improved baselines with momentum contrastive learning[J].arXiv:200304297,2020.
[31]GIDARIS S,BURSUC A,PUY G,et al.Obow:Online bag-of-visual-words generation for self-supervised learning[C]//Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.IEEE,2021:6826-6836.
[32]LIN X,LI C T.On constructing A better correlation predictor for PRNU-based image forgery localization[C]//Proceedings of 2021 IEEE International Conference on Multimedia and Expo.IEEE,2021:1-6.
[33]FERNÁNDEZ-MENDUIÑA S,PÉREZ-GONZÁLEZ F.On the information leakage quantification of camera fingerprint estimates[J].EURASIP Journal on Information Security,2021(1).
[34]PÉREZ-GONZÁLEZ F,FERNÁNDEZ-MENDUIÑA S.PRNU- leaks:Facts and remedies[C]//Proceedings of 2020 28th European Signal Processing Conference.2021:720-724.
[35]TIAN H,XIAO Y,CAO G,et al.Daxing smartphone identification dataset[J].IEEE Access,2019,7:101046-101053.
[36]AKBARI Y,ALMAADEED N,AL-MAADEED S,et al.PRNU-Net:A deep learning approach for source camera model identification based on videos taken with smartphone[C]//Proceedings of 2022 26th International Conference on Pattern Recognition.2022:599-605.
[37]ULYANOV D,VEDALDI A,LEMPITSKY V.Deep image prior[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.IEEE,2018:9446-9454.
[38]CHEN M,FRIDRICH J,GOLJAN M,et al.Source digital camcorder identification using sensor photo response non-uniformity[C]//Security,Steganography,and Watermarking of Multimedia Contents IX.2007:517-528.
[39]GLOE T,BÖHME R.The Dresden Image Database for benchmarking digital image forensics[C]//Proceedings of the 2010 ACM symposium on applied computing.New York:ACM,2010:1584-1590.
[40]XIAO Y,TIAN H,CAO G,et al.Effective PRNU extraction via densely connected hierarchical network[J].Multimedia Tools and Applications,2022,81(15):20443-20463.
[41]BERTINI F,SHARMA R,MONTESI D.Are social networks watermarking us or are we(unawarely) watermarking ourself?[J].Journal of Imaging,2022,8(5):132.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发