SRU-NET: SOBEL RESIDUAL U-NET FOR IMAGE MANIPULATION DETECTION

  • Baoxiang Jiang Xiamen University, Xiamen, China
  • Jingbo Xia Xiamen University Tan Kah Kee College, Zhangzhou, China
  • Yanting Wang Xiamen University, Xiamen, China
Keywords: Image Manipulation Detection, Sobel Residual, Central Differential Convolution

Abstract

Recently, most successful image manipulation detection methods have been based on convolutional neural networks (CNNs). Nevertheless, Existing CNN methods have limited abilities. CNN-based detection networks tend to extract signal features strongly related to content. However, image manipulation detection tends to extract weak signal features that are weakly related to content. To address this issue, We propose a novel Sobel residual neural network with adaptive central difference convolution, an extension of the classical U-Net architecture, for image manipulation detection. Adaptive central differential convolution can capture the essential attributes of an image by gathering intensity and gradient information. Sobel residual gradient block can capture forgery edge discriminative details. Extensive experimental results show that our method can significantly improve the accuracy of localising the forged region compared with the state-of-the-art methods.

Downloads

Download data is not yet available.

References

T. Bianchi, A. De Rosa, and A. Piva. Improved duct coefficient analysis for forgery localisation in jpeg images. InICASSP, 2011. 2, 3.
[2]. M. Goljan and J. Fridrich. Cfa-aware features for steganalysis of colour images. In SPIE/IS&T Electronic Imaging, 2015.1, 3.
[3]. D. Cozzolino, G. Poggi, and L. Verdoliva. Splicebuster: A new blind image splicing detector. In WIFS, 2015.
[4]. Fridrich, J., Soukal, D. and Lukas, J. (2003) Detection of Copy-Move Forgery in Digital Images. Proceedings of Digital Forensic Research Workshop, Cleveland, August 2003, 55-61.
[5]. Y. Rao and J. Ni. A deep learning approach to detection of splicing and copy-move forgeries in images. In WIFS, 2016.3.
[6]. ZHANG Ying, GOH J, WIN L Shukla, et al. Image region forgery detection: A Deep Learning Approach[M]. MATHUR A and ROY CHOUDHURY R. Proceedings of the Singapore Cyber-Security Conference. Amsterdam: IOS Press, 2016: 1–11.
[7]. Bappy M J H, Roy-Chowdhury A K, Bunk J, et al. Exploiting Spatial Structure for Localising Manipulated Image Regions[C]// International Conference on Computer Vision (ICCV), 2017. IEEE Computer Society, 2017.
[8]. Zhou, Peng, Han, Xintong, Morariu, Vlad I. Learning Rich Features for Image Manipulation Detection[J].
[9]. Y. Wu, W. AbdAlmageed, and P. Natarajan, "Mantra net: Manipulation tracing network for detection and localisation of image forgeries with anomalous features," in CVPR, 2019, pp. 9543–9552.
[10]. B. Mahdian and S. Saic, "Using noise inconsistencies for blind image forensics," Image and Vision Computing, vol. 27, no. 10, pp. 1497–1503, 2009.
[11]. Yu, Zitong, et al. "Searching central difference convolutional networks for face anti-spoofing." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.
[12]. J. Dong, W. Wang, and T. Tan, "CASIA image tampering detection evaluation database," in ChinaSIP, 2013, pp. 422–426.
[13]. Columbia image splicing detection evaluation dataset," http://www.ee.columbia.edu/ln /dvmm/ downlod /AuthSpliced DataSet /AuthSpliced DataSet.htm,2004.
[14]. "Nist 2016 datasets," https://www. nist.gov/sites/ default /files/ documents /2016/11/30/should_i_believe or not.pdf.
[15]. B. Wen, Y. Zhu, R. Subramanian, T. Ng, X. Shen, and S. Winkler, "Coverage a novel database for copy-move forgery detection," in ICIP, Sep. 2016, pp. 161–165.
[16]. Krawetz, N., Solutions, HF: A picture's worth. Hacker Factor Solutions 6(2), 2(2007)
[17]. R. Salloum, Y. Ren, and C.-C. J. Kuo. Image splicing localisation using a multi-task fully convolutional network(mfcn). arXiv preprint arXiv:1709.02016, 2017. 3, 6.
[18]. J. H. Bappy, A. K. Roy-Chowdhury, J. Bunk, L. Nataraj, and B. Manjunath. Exploiting spatial structure for localising manipulated image regions. In ICCV, 2017. 1, 3, 6.
[19]. M. Zampoglou, S. Papadopoulos, and Y. Kompatsiaris, "Large-scale evaluation of splicing localisation algorithms for web images," Multimedia Tools Appl., vol.76, no. 4, pp. 4801–4834, 2017.
[20]. O. Ronneberger, P. Fischer, and T. Brox. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, pages 234–241. Springer, 2015.
[21]. K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages770–778, 2016.
[22]. Arvind Mewada, and Rupesh Kumar Dewang. "Research on False Review Detection Methods: A state-of-the-art review." Journal of King Saud University-Computer and Information Sciences (2021).
Published
2021-09-14
How to Cite
Jiang, B., Xia, J., & Wang, Y. (2021). SRU-NET: SOBEL RESIDUAL U-NET FOR IMAGE MANIPULATION DETECTION. International Journal of Advanced Computer Technology, 10(5), 01-06. Retrieved from https://ijact.org/index.php/ijact/article/view/89