Full Text PDF
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Vol 10 No 5 (2026): Volume 10, Issue 5, May 2026 | Pages: 345-352
International Research Journal of Innovations in Engineering and Technology
OPEN ACCESS | Research Article | Published Date: 17-05-2026
In the modern digital era, automated bots and malicious scripts have become a major threat to online platforms. Traditional CAPTCHA systems are increasingly vulnerable to machine learning attacks. This project proposes an Adversarial CAPTCHA Security System using Generative Adversarial Networks (GANs) and deep learning techniques to improve CAPTCHA robustness against automated bots. The system generates adversarial CAPTCHA images by adding GAN generated perturbations to real images from the CIFAR-10 dataset. These adversarial images remain understandable for humans while making automated recognition difficult for bots. The system also includes secure user authentication using Flask and SQLite database integration. Experimental results demonstrate improved CAPTCHA security and enhanced resistance against automated attacks.
GAN, CAPTCHA, Deep Learning, CIFAR-10, Adversarial Images, Flask, SQLite.
Vinayak Patil, Punit Patil, Lokesh Patil, Dinesh Patil, & Piyush Patil. (2026). Secure CAPTCHA with Patch Base Defense. International Research Journal of Innovations in Engineering and Technology - IRJIET, 10(5), 345-352. Article DOI https://doi.org/10.47001/IRJIET/2026.105045
This work is licensed under Creative common Attribution Non Commercial 4.0 Internation Licence
CIFAR-10 Dataset – Learning Multiple Layers of Features from Tiny Images. Alex Krizhevsky, 2009.
PyTorch Official Documentation – https://pytorch.org/docs/
Flask Web Framework Documentation – https://flask.palletsprojects.com/
Ian Goodfellow, Jonathon Shlens, Christian Szegedy – Explaining and Harnessing Adversarial Examples. ICLR, 2015.
Luis von Ahn, Manuel Blum, Nicholas Hopper, John Langford – CAPTCHA: Using Hard AI Problems for Security. EUROCRYPT, 2003.
A.Nguyen, J. Yosinski, and J. Clune, “Deep neural networks are easily fooled: High confidence predictions for unrecognizable images,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 427–436.
T. B. Brown, D. Man’s, A. Roy, M. Abadi, and J. Gilmer, “Adversarial patch,” arrive preprint arXiv:1712.09665, 2017.
C. Szegedy, W. Zaremba, I. Stuever, J. Bruna, D. Erhan, I. Goodfellow, and R. Fergus, “Intriguing properties of neural networks,” in International Conference on Learning Representations (ICLR), 2014.
N. Papernot, P. McDaniel, and I. Goodfellow, “Transferability in machine learning: From phenomena to black-box attacks using adversarial samples,” arrive preprint arXiv:1605.07277, 2016.
F. Trammer, A. Kurakin, N. Papernot, D. Bone, and P. D. McDaniel, “Ensemble adversarial training: Attacks and defenses,” arrive preprint arXiv:1705.07204, 2017.
C. Shi, X. Xu, S. Ji, K. Bu, J. Chen, R. A. Beyah, and T. Wang, “Adversarial CAPTCHAs,” arrive preprint arXiv:1901.01107, 2019.