Abstract
In this paper, we propose a strategy to train a CNN to detect document manipulations in JPEG documents under data scarcity scenario. As it comes to scanned PDF documents, it is common that the document consists of a JPEG image encapsulated into a PDF. Indeed, if the document before tampering was a JPEG image, its manipulation will lead to double compression artefacts within the resulting tampered JPEG image. In contrast to related methods that are based on handcrafted histograms of DCT coefficients, we propose a double compression detection method using a one-hot encoding of the DCT coefficients of JPEG images. We can use accordingly a CNN model to compute co-occurrence matrices and avoid handcrafted features such as histograms. Using simulated frauds on Perlin noise, we train our network and then test it on textual images against a state-of-the-art CNN algorithm trained on natural images. Our approach has shown an encouraging generalization on both the database used in the paper and on a stream of synthetic frauds on real documents used in the company Yooz.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Perlin noise is a procedural texture primitive, it is a gradient noise used to improve the realism of the CGI.
References
Artaud, C., Sidère, N., Doucet, A., Ogier, J.M., Yooz, V.P.D.: Find it! fraud detection contest report. In: 2018 24th International Conference on Pattern Recognition (ICPR), pp. 13–18. IEEE (2018)
Bae, H.J., et al.: A Perlin noise-based augmentation strategy for deep learning with small data samples of HRCT images. Sci. Rep. 8(1), 1–7 (2018)
Barni, M., et al.: Aligned and non-aligned double jpeg detection using convolutional neural networks. J. Vis. Commun. Image Represent. 49, 153–163 (2017)
Bianchi, T., Piva, A.: Analysis of non-aligned double jpeg artifacts for the localization of image forgeries. In: 2011 IEEE International Workshop on Information Forensics and Security, pp. 1–6. IEEE (2011)
Chen, Y.L., Hsu, C.T.: Detecting recompression of JPEG images via periodicity analysis of compression artifacts for tampering detection. IEEE Trans. Inf. Forensics Secur. 6(2), 396–406 (2011)
Farid, H.: Exposing digital forgeries from JPEG ghosts. IEEE Trans. Inf. Forensics Secur. 4(1), 154–160 (2009)
Fu, D., Shi, Y.Q., Su, W.: A generalized Bedford’s law for JPEG coefficients and its applications in image forensics. In: Security, Steganography, and Watermarking of Multimedia Contents IX. vol. 6505, p. 65051L. International Society for Optics and Photonics (2007)
Huang, J.: Segmentation-based hybrid compression scheme for scanned documents (May 20 2008), US Patent 7,376,265
Khermaza, E., Tkachenko, I., Picard, J.: Can copy detection patterns be copied? evaluating the performance of attacks and highlighting the role of the detector. In: 2021 IEEE International Workshop on Information Forensics and Security (WIFS), pp. 1–6. IEEE (2021)
Kwon, M.J., Yu, I.J., Nam, S.H., Lee, H.K.: Cat-net: Compression artifact tracing network for detection and localization of image splicing. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 375–384 (2021)
Li, B., Shi, Y.Q., Huang, J.: Detecting doubly compressed jpeg images by using mode based first digit features. In: 2008 IEEE 10th Workshop on Multimedia Signal Processing, pp. 730–735. IEEE (2008)
Lin, Z., He, J., Tang, X., Tang, C.K.: Fast, automatic and fine-grained tampered JPEG image detection via DCT coefficient analysis. Pattern Recogn. 42(11), 2492–2501 (2009)
Park, J., Cho, D., Ahn, W., Lee, H.-K.: Double JPEG detection in mixed JPEG quality factors using deep convolutional neural network. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11209, pp. 656–672. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01228-1_39
Popescu, A.C., Farid, H.: Statistical tools for digital forensics. In: Fridrich, J. (ed.) IH 2004. LNCS, vol. 3200, pp. 128–147. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-30114-1_10
van Renesse, R.L.: Paper based document security-a review. In: European Conference on Security and Detection, 1997. ECOS 1997, pp. 75–80. IET (1997)
Umesh, P.: Image processing in python. CSI Commun. 23, 1–25 (2012)
Wang, Q., Zhang, R.: Double JPEG compression forensics based on a convolutional neural network. EURASIP J. Inf. Secur. 2016(1), 1–12 (2016)
Wu, L., et al.: Editing text in the wild. In: Proceedings of the 27th ACM International Conference on Multimedia, pp. 1500–1508 (2019)
Yousfi, Y., Fridrich, J.: An intriguing struggle of CNNs in JPEG steganalysis and the onehot solution. IEEE Signal Process. Lett. 27, 830–834 (2020)
Yu, F., Koltun, V.: Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122 (2015)
Yu, I.J., Nam, S.H., Ahn, W., Kwon, M.J., Lee, H.K.: Manipulation classification for JPEG images using multi-domain features. IEEE Access 8, 210837–210854 (2020)
Acknowledgments
This work is supported by the Region Nouvelle Aquitaine under the grant number 2019-1R50120 (CRASD project) and AAPR2020-2019-8496610 (CRASD2 project) and by the LabCom IDEAS under the grant number ANR-18-LCV3-0008.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 Springer Nature Switzerland AG
About this paper
Cite this paper
Taburet, T. et al. (2023). Document Forgery Detection in the Context of Double JPEG Compression. In: Rousseau, JJ., Kapralos, B. (eds) Pattern Recognition, Computer Vision, and Image Processing. ICPR 2022 International Workshops and Challenges. ICPR 2022. Lecture Notes in Computer Science, vol 13646. Springer, Cham. https://doi.org/10.1007/978-3-031-37745-7_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-37745-7_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-37744-0
Online ISBN: 978-3-031-37745-7
eBook Packages: Computer ScienceComputer Science (R0)