Skip to main content
SpringerLink
Log in

A novel general blind detection model for image forensics based on DNN

  • Original article
  • Published:
The Visual Computer Aims and scope Submit manuscript

Abstract

Image steganography and image tampering usually produce weak characteristic signals that are different from the natural information of the image. Aiming at the unnatural features in images, this paper proposes a blind detection model for image forensics based on weak feature extraction. The model extracts weak image features from three aspects: the spatial domain, the JPEG domain, and the natural characteristics. It has the characteristics of a wide detection range and a good detection accuracy. The model can perform general blind detection for image spatial domain steganography, image JPEG domain steganography, image copy-move, image splicing, and image removal. It is mainly composed of four types of artificial neural network modules, two feature classification networks, and a target regression network. The model combines the multi-layer convolution RoI feature extraction method and uses three deep residual networks and RPN to extract weak features of the image. The model integrates the features extracted by the four networks and makes the final judgment on the image detection through two feature classification networks and a target regression network. We tested the image steganography algorithms in the typical spatial domain and JPEG domain, and image content tampering operations such as image copy-move, splicing, and removal. At the same time, we made a double image dataset containing tampering and steganography and tested the model’s ability to detect the double image dataset. The experimental results show that the model has a relatively ideal detection effect on the typical algorithms tested. The model can also detect the mixed steganography and tampering information in the double image dataset.

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
Fig. 5
Fig. 6

Similar content being viewed by others

References

  1. Holub, V., Fridrich, J., Denemark, T.: Universal distortion function for steganography in an arbitrary domain. Eurasip J. Inf. Sec. 2014(1), 1 (2014)

    Article  Google Scholar 

  2. Li, B., Ming, W., Huang, J., Li, X.: A new cost function for spatial image steganography. In: 2014 IEEE international conference on image processing (ICIP) (2015)

  3. Holub, V., Fridrich, J.: Designing steganographic distortion using directional filters. In: IEEE workshop on information forensic and security (2012)

  4. Guo, L., Ni, J., Shi, Y.Q.: Uniform embedding for efficient jpeg steganography. IEEE Trans. Inf. Forensics Sec. 9(5), 814–825 (2014). https://doi.org/10.1109/TIFS.2014.2312817

    Article  Google Scholar 

  5. Memon, N.D., Sankur, B.: Steganalysis of watermarking techniques using image quality metrics. Proc. Spie Int. Soc. Opt. Eng. 4314, 523–531 (2001)

    Google Scholar 

  6. Kharrazi, M., Memon, M.: Image steganalysis with binary similarity measures. EURASIP J. Adv. Signal Process. 17, 1–9 (2005)

    MATH  Google Scholar 

  7. Lyu, S., Farid, H.: Detecting hidden messages using higher-order statistics and support vector machines. Lect. Notes Comput. Sci. 2578, 340–354 (2003)

    Article  MATH  Google Scholar 

  8. Zhang, L., Yan, Q., Zhu, Y., Zhang, X., Xiao, C.: Effective shadow removal via multi-scale image decomposition. Vis. Comput. Int. J. Comput. Graph. 35(6–8), 1091–1104 (2019)

    Google Scholar 

  9. Joseph, A., Geetha, P.: Facial emotion detection using modified eyemap-mouthmap algorithm on an enhanced image and classification with tensorflow. Vis. Comput. 36(3), 529–539 (2019)

    Article  Google Scholar 

  10. Guclu, O., Can, A.B.: Integrating global and local image features for enhanced loop closure detection in rgb-d slam systems. Vis. Comput. 36(5), 1271–1290 (2020)

    Article  Google Scholar 

  11. Ran, M., Zelnik-Manor, L., Tal, A.: Saliency for image manipulation. Vis. Comput. 29(5), 381–392 (2013)

    Article  Google Scholar 

  12. Wu, Y., Abdalmageed, W., Natarajan, P.: Mantra-net: Manipulation tracing network for detection and localization of image forgeries with anomalous features. In: 2019 IEEE/CVF conference on computer vision and pattern recognition (CVPR) (2019)

  13. Hu, X., Zhang, Z., Jiang, Z., Chaudhuri, S., Yang, Z., Nevatia, R.: Span: Spatial pyramid attention network forimage manipulation localization (2020)

  14. Chen, X., Dong, C., Ji, J., Cao, J., Li, X.: Image manipulation detection by multi-view multi-scale supervision (2021)

  15. Liu, X., Liu, Y., Chen, J., Liu, X.: Pscc-net: Progressive spatio-channel correlation network for image manipulation detection and localization (2021)

  16. Zhong, Z., Jin, L., Huang, S.: Deeptext: A unified framework for text proposal generation and text detection in natural images. In: IEEE international conference on acoustics, pp. 1208–1212 (2016)

  17. Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks. IEEE Trans. Pattern Anal. Mach. Intell. 39(6), 1137–1149 (2017)

    Article  Google Scholar 

  18. Petitcolas, F., Anderson, R.J., Kuhn, M.G.: Information hiding-a survey. Proc. IEEE 87(7), 1062–1078 (1999)

    Article  Google Scholar 

  19. Provos, N.: Defending against statistical steganalysis. In: Proceedings of 10th USENIX Security Symposium, 2001 (2001)

  20. Westfeld, A.: F5-a steganographic algorithm. In: proceedings of the 4th international workshop information hiding, IHW 2001, Pittsburgh, PA, USA, April pp. 25-27, (2001)

  21. Sallee, P.: Model-based steganography. In: international workshop on digital watermarking (2003)

  22. Qian, Y., Dong, J., Wang, W., Tan, T.: Deep learning for steganalysis via convolutional neural networks. In: conference on media watermarking, security, and forensics (2015)

  23. Xu, G., Wu, H., Shi, Y.: Ensemble of cnns for steganalysis: an empirical study. pp. 103–107 (2016). doi: https://doi.org/10.1145/2909827.2930798

  24. Xu, G., Wu, H.Z., Shi, Y.Q.: Structural design of convolutional neural networks for steganalysis. IEEE Signal Process. Lett. 23(5), 708–712 (2016)

    Article  Google Scholar 

  25. Jian, Y., Ni, J., Yang, Y.: Deep learning hierarchical representations for image steganalysis. IEEE Trans. Inf. Forensics Sec. 12(11), 2545–2557 (2017)

    Article  Google Scholar 

  26. Fridrich, J.: Rich models for steganalysis of digital images. IEEE Trans. Inf. Forensics Sec. 7(3), 868–882 (2012)

    Article  Google Scholar 

  27. Boroumand, M., Chen, M., Fridrich, J.: Deep residual network for steganalysis of digital images. IEEE Trans. Inf. Forensics Sec. 14(5), 1181–1193 (2018)

    Article  Google Scholar 

  28. Bunk, J., Bappy, J.H., Mohammed, T.M., Nataraj, L., Flenner, A., Manjunath, B.S., Chandrasekaran, S., Roy-Chowdhury, A.K., Peterson, L.: Detection and localization of image forgeries using resampling features and deep learning. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2017)

  29. Bappy, M., Roy-Chowdhury, A.K., Bunk, J., Nataraj, L., Manjunath, B.S.: Exploiting spatial structure for localizing manipulated image regions. In: IEEE international conference on computer vision (2017)

  30. Park, J., Cho, D., Ahn, W., Lee, H.K.: Double JPEG Detection in mixed JPEG quality factors using deep convolutional neural network. In: Proceedings 15th European Conference, Munich, Germany, September 8–14, 2018, Part V. Computer Vision - ECCV 2018,(2018)

  31. Zhou, P., Han, X., Morariu, V.I., Davis, L.S.: Learning rich features for image manipulation detection. In: 2018 IEEE/CVF conference on computer vision and pattern recognition (CVPR) (2018)

  32. Bappy, M.J.H., Simons, C., Nataraj, L., Manjunath, B.S., Roy-Chowdhury, A.K.: Hybrid lstm and encoder-decoder architecture for detection of image forgeries. IEEE Trans. Image Process. 28(7), 3286–3300 (2019)

    Article  MathSciNet  MATH  Google Scholar 

  33. Wei, X., Wu, Y., Dong, F., Zhang, J., Sun, S.: Developing an image manipulation detection algorithm based on edge detection and faster r-cnn. Symmetry (2019). https://doi.org/10.3390/sym11101223

    Article  Google Scholar 

  34. Marr, D., Hildreth, E.: Theory of edge detection. Proc. R. Soc. Lond. 207(1167), 187–217 (1980)

    Google Scholar 

  35. Prewitt: object enhancement and extraction (1970)

  36. Nist nimble 2016 datasets. https://www.nist.gov/itl/iad/mig/nimble-challenge-2017-evaluation/

  37. Bas, P., Filler, T., Pevn, T.: Break our steganographic system: the ins and outs of organizing boss. Information Hiding. In: 13th international conference, IH 2011, Prague, Czech Republic, May 18-20, 2011, Revised Selected Papers (2011)

  38. Bas, P., Furon, T.: Bows-2 (2007). http://bows2.ec-lille.fr

  39. Sedighi, V., Cogranne, R., Fridrich, J.: Content-adaptive steganography by minimizing statistical detectability. IEEE Trans. Inf. Forensics Sec. 11(2), 1–1 (2015)

    Google Scholar 

  40. Lin, T.Y., Maire, M., Belongie, S., Hays, J., Zitnick, C.L.: Microsoft coco: common objects in context. Springer, Berlin (2014)

    Google Scholar 

  41. Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M.: Imagenet large scale visual recognition challenge. Int. J. Comput. Vis. 115(3), 211–252 (2014)

    Article  MathSciNet  Google Scholar 

  42. Dong, J., Wang, W., T, T.: Casia image tampering detection evaluation database 2010. http://forensics.idealtest.org

  43. Dong, J., Wang, W., T, T.: Casia image tampering detection evaluation database (2013)

  44. Columbia uncompressed image splicing detection evaluation dataset. http://www.ee.columbia.edu/ln/dvmm/downloads/authsplcuncmp/

  45. Casia 1.0 groundtruth. https://github.com/namtpham/casia1groundtruth

  46. Casia 2.0 groundtruth. https://github.com/namtpham/casia2groundtruth

  47. Xu, G.: Deep convolutional neural network to detect J-UNIWARD. In: proceedings of the 5th ACM workshop on information hiding and multimedia security, IH MMSec 2017, Philadelphia, PA, USA, June 20-22, 2017, pp. 67–73. ACM (2017). https://doi.org/10.1145/3082031.3083236

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China [Grant Numbers 61771168]. The authors would like to thank the Institute of Information Countermeasures Technology providing deep learning servers.

Author information

Authors and Affiliations

  1. School of Computer Science and Technology, Harbin Institute of Technology, Harbin, Heilongjiang, 150001, China

    Hao Chen, Qi Han, Qiong Li & Xiaojun Tong

Authors
  1. Hao Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Qi Han
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Qiong Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Xiaojun Tong
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Qi Han.

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

Chen, H., Han, Q., Li, Q. et al. A novel general blind detection model for image forensics based on DNN. Vis Comput 39, 27–42 (2023). https://doi.org/10.1007/s00371-021-02310-3

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00371-021-02310-3

Keywords

Search

Navigation