Publication Details
Keywords: Certificate Forgery Detection, Convolutional Neural Network, Grad-CAM, Deep Learning, Document Verification, Binary Classification.
Abstract
Government-issued caste certificates serve as a gateway to constitutional benefits reserved for socially and economically disadvantaged communities in India. Growing misuse of image manipulation tools to fabricate such documents has created an urgent need for reliable automated verification mechanisms. This study introduces a deep-learning solution, referred to as LOCATE, that performs both classification and spatial fault-finding on suspect certificates. A four-layer Convolutional Neural Network trained on a purpose-built collection of 1,000 Karnataka state certificates (500 genuine, 500 manipulated) first separates authentic documents from tampered ones. Gradient-weighted Class Activation Mapping then pinpoints the exact region within a flagged document that triggered the forgery verdict, giving human reviewers a clear visual basis for their decisions. At operating threshold τ = 0.7, the system attains 98% overall accuracy, 100% precision, 96% recall, and an F1-score of 0.98, while preserving 100% specificity. These results surpass all comparable prior approaches and demonstrate that the framework is ready for real-world administrative deployment.
References
- [1] V. Christlein, C. Riess, J. Jordan, C. Riess, and E. Angelopoulou, "An evaluation of popular copy-move forgery detection approaches," IEEE Trans. Inf. Forensics Security, vol. 7, no. 6, pp. 1841–1854, Dec. 2012. [2] S.-J. Ryu, M.-J. Lee, and H.-K. Lee, "Detection of copy-rotate-move forgery using Zernike moments," in Proc. Int. Workshop Inf. Hiding, Calgary, Canada, 2010, pp. 51–65. [3] X. Bi, C.-M. Pun, and X.-C. Yuan, "Multi-scale location-aware kernel representation for object detection," in Proc. IEEE/CVF CVPR, Salt Lake City, UT, 2018, pp. 1248–1257. [4] R. Salloum, Y. Ren, and C.-C. J. Kuo, "Image splicing localization using a multi-task fully convolutional network (MFCN)," J. Vis. Commun. Image Represent., vol. 51, pp. 201–209, Feb. 2018. [5] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, "Grad-CAM: Visual explanations from deep networks via gradient-based localization," in Proc. IEEE ICCV, Venice, Italy, 2017, pp. 618–626. [6] B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, "Learning deep features for discriminative localization," in Proc. IEEE/CVF CVPR, Las Vegas, NV, 2016, pp. 2921–2929. [7] P. Joshi, A. Bhatt, and A. Patel, "Detecting forgery in PAN cards using local binary patterns and support vector machines," Int. J. Comput. Appl., vol. 183, no. 14, pp. 1–6, Jul. 2021. [8] A. Kumar, S. Singh, and R. Gupta, "Aadhaar card forgery detection using deep convolutional features," J. Inf. Security Appl., vol. 62, p. 103010, Oct. 2021. [9] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, pp. 436–444, May 2015. [10] D. P. Kingma and J. Ba, "Adam: A method for stochastic optimization," in Proc. ICLR, San Diego, CA, 2015, arXiv:1412.6980. [11] S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," in Proc. ICML, Lille, France, 2015, pp. 448–456. [12] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: A simple way to prevent neural networks from overfitting," J. Mach. Learn. Res., vol. 15, pp. 1929–1958, 2014. [13] K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," in Proc. ICLR, San Diego, CA, 2015. [14] O. Russakovsky et al., "ImageNet large-scale visual recognition challenge," Int. J. Comput. Vis., vol. 115, no. 3, pp. 211–252, 2015.