Skip to main content
SpringerLink
Log in

Deep siamese network for limited labels classification in source camera identification

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

Source Camera Identification is a well-known digital forensic challenge of mapping an image to its authentic source. The current state-of-the-art provides a number of successful and efficient solutions to this problem. However, in almost all such existing techniques, a sufficiently large number of image samples is required for pre-processing, before source identification. Limited labels classification is a realistic scenario for a forensic analyst where s/he has access only to a few labelled training samples, available for source camera identification. In such contexts, where obtaining a vast number of image samples (per camera) is infeasible, correctness of existing source identification schemes, is threatened. In this paper, we address the problem of performing accurate source camera identification, with a limited set of labelled training samples, per camera model. We use a few shot learning technique known as deep siamese network here, and achieve significantly improved classification accuracy than the state–of–the–art. Here, the main principle of operation is to form pairs of samples from the same camera models, as well as from different camera models, to enhance the training space. Subsequently, a deep neural network is used to perform source classification. We perform experiments on traditional camera model identification, as well as intra–make and intra–device source identification. We also show that our proposed methodology under limited labels scenario, is robust to image transformations such as rotation, scaling, compression, and additive noise.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. The imnoise() function of MATLAB is used in our work for additional noise.

References

  1. Akshatha K, Karunakar A, Anitha H, Raghavendra U, Shetty D (2016) Digital camera identification using prnu: a feature based approach. Digit Investig 19:69–77

    Article  Google Scholar 

  2. Bayram S, Sencar HT, Memon N (2015) Sensor fingerprint identification through composite fingerprints and group testing. IEEE Trans Inf Forensics Secur 10(3):597–612

    Article  Google Scholar 

  3. Berlemont S, Lefebvre G, Duffner S, Garcia C (2018) Class-balanced siamese neural networks. Neurocomputing 273:47–56

    Article  Google Scholar 

  4. Bondi L, Baroffio L, Güera D, Bestagini P, Delp EJ, Tubaro S (2017) First steps toward camera model identification with convolutional neural networks. IEEE Signal Process Lett 24(3):259– 263

    Article  Google Scholar 

  5. Ċeliktutan O, Sankur B, et al. (2008) Blind identification of source cell-phone model. IEEE Trans Inf Forensics Secur 3(3):553–566

    Article  Google Scholar 

  6. Chen C, Stamm MC (2015) Camera model identification framework using an ensemble of demosaicing features. In: Information forensics and security (WIFS), 2015 IEEE international workshop on. IEEE, pp 1–6

  7. Chen M, Fridrich J, Goljan M, Lukás J (2008) Determining image origin and integrity using sensor noise. IEEE Trans Inf Forensics Secur 3 (1):74–90

    Article  Google Scholar 

  8. Chopra S, Hadsell R, LeCun Y (2005) Learning a similarity metric discriminatively, with application to face verification. In: Computer vision and pattern recognition, 2005. CVPR 2005. IEEE computer society conference on, vol 1. IEEE, pp 539–546

  9. Dirik AE, Sencar HT, Memon N (2008) Digital single lens reflex camera identification from traces of sensor dust. IEEE Trans Inf Forensics Secur 3(3):539–552

    Article  Google Scholar 

  10. Dirik AE, Sencar HT, Memon N (2014) Analysis of seam-carving-based anonymization of images against prnu noise pattern-based source attribution. IEEE Trans Inf Forensics Secur 9(12):2277– 2290

    Article  Google Scholar 

  11. Fei-Fei L, Fergus R, Perona P (2006) One-shot learning of object categories. IEEE Trans Pattern Anal Mach Intell 28(4):594–611

    Article  Google Scholar 

  12. Gloe T (2012) Feature-based forensic camera model identification. In: Transactions on data hiding and multimedia security VIII. Springer, pp 42–62

  13. Gloe T, Böhme R (2010) The dresden image database for benchmarking digital image forensics. J Digit Forensic Prac 3(2-4):150–159

    Article  Google Scholar 

  14. Goljan M, Fridrich J, Filler T (2009) Large scale test of sensor fingerprint camera identification. In: Media forensics and security. International society for optics and photonics, vol 7254, p 72540i

  15. Hadsell R, Chopra S, LeCun Y (2006) Dimensionality reduction by learning an invariant mapping. In: Computer vision and pattern recognition. 2006 IEEE computer society conference on, vol 2. IEEE, pp 1735–1742

  16. Huang Y, Zhang J, Huang H (2015) Camera model identification with unknown models. IEEE Trans Inf Forensics Secur 10(12):2692–2704

    Article  Google Scholar 

  17. Karaküċük A, Dirik AE (2015) Adaptive photo-response non-uniformity noise removal against image source attribution. Digit Investig 12:66–76

    Article  Google Scholar 

  18. Kharrazi M, Sencar HT, Memon N (2004) Blind source camera identification. In: Image processing, 2004. ICIP’04. 2004 international conference on, vol 1. IEEE, pp 709–712

  19. Koch G (2015) Siamese neural networks for one-shot image recognition. In: 32nd international conference on machine learning. International machine learning society, vol 37

  20. Lawgaly A, Khelifi F (2017) Sensor pattern noise estimation based on improved locally adaptive dct filtering and weighted averaging for source camera identification and verification. IEEE Trans Inf Forensics Secur 12(2):392–404

    Article  Google Scholar 

  21. Li CT (2010) Source camera identification using enhanced sensor pattern noise. IEEE Trans Inf Forensics Secur 5(2):280–287

    Article  Google Scholar 

  22. Lin X, Li CT (2016) Preprocessing reference sensor pattern noise via spectrum equalization. IEEE Trans Inf Forensics Secur 11(1):126–140

    Article  Google Scholar 

  23. Lin X, Li CT (2017) Large-scale image clustering based on camera fingerprints. IEEE Trans Inf Forensics Secur 12(4):793–808

    Google Scholar 

  24. Lukas J, Fridrich J, Goljan M (2006) Digital camera identification from sensor pattern noise. IEEE Trans Inf Forensics Secur 1(2):205–214

    Article  Google Scholar 

  25. Marra F, Poggi G, Sansone C, Verdoliva L (2016) Correlation clustering for prnu-based blind image source identification. In: Information forensics and security (WIFS). 2016 IEEE international workshop on. IEEE, pp 1–6

  26. Marra F, Poggi G, Sansone C, Verdoliva L (2017) Blind prnu-based image clustering for source identification. IEEE Trans Inf Forensics Secur 12(9):2197–2211

    Article  Google Scholar 

  27. Marra F, Poggi G, Sansone C, Verdoliva L (2017) A study of co-occurrence based local features for camera model identification. Multimed Tools Appl 76(4):4765–4781

    Article  Google Scholar 

  28. Sabri M, Kurita T (2018) Facial expression intensity estimation using siamese and triplet networks. Neurocomputing 313:143–154

    Article  Google Scholar 

  29. Shullani D, Fontani M, Iuliani M, Al Shaya O, Piva A (2017) Vision: a video and image dataset for source identification. EURASIP J Inf Secur 2017(1):15

    Article  Google Scholar 

  30. Tan Y, Wang B, Li M, Guo Y, Kong X, Shi Y (2015) Camera source identification with limited labeled training set. In: International workshop on digital watermarking. Springer, pp 18–27

  31. Thai TH, Retraint F, Cogranne R (2015) Camera model identification based on dct coefficient statistics. Digit Signal Process 40:88–100

    Article  MathSciNet  Google Scholar 

  32. Thai TH, Retraint F, Cogranne R (2016) Camera model identification based on the generalized noise model in natural images. Digit Signal Process 48:285–297

    Article  MathSciNet  Google Scholar 

  33. Tuama A, Comby F, Chaumont M (2015) Source camera model identification using features from contaminated sensor noise. In: International workshop on digital watermarking. Springer , pp 83–93

  34. Tuama A, Comby F, Chaumont M (2016) Camera model identification with the use of deep convolutional neural networks. In: IEEE International workshop on information forensics and security , pp 6–pages

  35. Tuama A, Comby F, Chaumont M (2016) Camera model identification with the use of deep convolutional neural networks. In: IEEE International workshop on information forensics and security, pp 6–pages

  36. Xu B, Wang X, Zhou X, Xi J, Wang S (2016) Source camera identification from image texture features. Neurocomputing 207:131–140

    Article  Google Scholar 

  37. Xu G, Shi YQ (2012) Camera model identification using local binary patterns. In: Multimedia and expo (ICME). 2012 IEEE international conference on. IEEE, pp 392–397

  38. Yang Y, Saleemi I, Shah M (2013) Discovering motion primitives for unsupervised grouping and one-shot learning of human actions, gestures, and expressions. IEEE Trans Pattern Anal Mach Intell 35(7):1635–1648

    Article  Google Scholar 

  39. Zeng H (2016) Rebuilding the credibility of sensor-based camera source identification. Multimed Tools Appl 75(21):13871–13882

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, National Institute of Technology, Rourkela, 769008, India

    Venkata Udaya Sameer

  2. Department of Information Technology, Indian Institute of Engineering Science and Technology, Shibpur, 711103, India

    Ruchira Naskar

Authors
  1. Venkata Udaya Sameer
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ruchira Naskar
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Venkata Udaya Sameer.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sameer, V.U., Naskar, R. Deep siamese network for limited labels classification in source camera identification. Multimed Tools Appl 79, 28079–28104 (2020). https://doi.org/10.1007/s11042-020-09106-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-020-09106-y

Keywords

Search

Navigation