Double-branch forgery image detection based on multi-scale feature fusion

Abstract

Most of existing methods exhibit poor performance in detecting forged images due to the small size of tampered areas and the limited pixel difference between untampered and tampered regions. To alleviate the above problem, a double-branch tampered image detection based on multi-scale features is proposed. Firstly, we introduce a fusion module based on attention mechanism in the first branch to enhance the network’s sensitivity towards tampered regions. Secondly, we construct a second branch specifically designed for detection, aiming to identify subtle differences between tampered and untampered areas by utilizing rich edge information from shallow features as guidance. Compared to the existing methods on the public benchmark datasets CASIA1.0, Columbia and NIST16, the values of F-score reached 0.766, 0.900 and 0.930 on those datasets, respectively. The experimental results show that our method could significantly improve the accuracy on detecting the tampered area.

References

[1] RaoY, NiJ. A deep learning approach to detection of splicing and copy-move forgeries in images. 2016 IEEE International Workshop on Information Forensics and Security (WIFS), December 5–7, 2016, Abu Dhabi, United Arab Emirates, 2016, New York, IEEE: 1-6[C]

[2] ZhouP, HanX, MorariuV I, et al.. Learning rich features for image manipulation detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 18–22, 2018, Salt Lake City, USA, 2018, New York, IEEE: 1053-1061[C]

[3] WangC, LiY, WuG. Image splicing tamper detection based on deep learning and attention mechanism. 2021 IEEE 6th International Conference on Signal and Image Processing (ICSIP), July 9–11, 2021, Nanjing, China, 2021, New York, IEEE: 267-271[C]

[4] ChenX, DongC, JiJ, et al.. Image manipulation detection by multi-view multi-scale supervision. Proceedings of the IEEE/CVF International Conference on Computer Vision, October 10–17, 2021, Montreal, Canada, 2021, New York, IEEE: 14185-14193[C]

[5] JiangX Y, LiuC X. Edge and region inconsistency-guided image splicing tamper detection network. Journal of image and graphics, 2021, 26(10):2411-2420[J]

[6] BiX L, WeiY, XiaoB, et al.. Image tamper detection algorithm based on cascaded convolutional neural network. Journal of electronics & information technology, 2019, 41(12):2987-2994[J]

[7] ChenL C, ZhuY, PapandreouG, et al.. Encoder-decoder with atrous separable convolution for semantic image segmentation. Proceedings of the European Conference on Computer Vision (ECCV), September 8–14, 2018, Munich, Germany, 2018, New York, IEEE: 801-818[C]

[8] ZhuH Y, SunJ, ChenQ D. Multi-task algorithm for image splicing forgery detection based on deepLab v3+. Computer engineering, 2022, 48(1):253-259[J]

[9] SandlerM, HowardA, ZhuM, et al.. Mobile-netv2: inverted residuals and linear bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 18–22, 2018, Salt Lake City, USA, 2018, New York, IEEE: 4510-4520[C]

[10] HuJ, ShenL, SunG. Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 18–22, 2018, Salt Lake City, USA, 2018, New York, IEEE: 7132-7141[C]

[11] DongJ, WangW, TanT. Casia image tampering detection evaluation database. 2013 IEEE China Summit and International Conference on Signal and Information Processing, July 6–10, 2013, Beijing, China, 2013, New York, IEEE: 422-426[C]

[12] BappyJ H, SimonsC, NatarajL, et al.. Hybrid LSTM and encoder-decoder architecture for detection of image forgeries. IEEE transactions on image processing, 2019, 28(7):3286-3300 J]

[13] HeK, GkioxariG, DollarP, et al.. Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision, October 24–27, 2017, Venice, Italy, 2017, New York, IEEE: 2961-2969[C]

[14] WeiX Y, ZuoX L, DanZ P, et al.. FCR-CNN model to improve the performance of image tamper detection region selection. Journal of computer-aided design & computer graphics, 2021, 33(4):560-568 J]