Enhancing Security of Online Interfaces: Adversarial Handwritten Arabic CAPTCHA Generation
Source abstract
Abstract
With the increasing online activity of Arabic speakers, the development of effective CAPTCHAs (Completely Automated Public Turing Tests to Tell Computers and Humans Apart) tailored for Arabic users has become crucial. Traditional CAPTCHAs, however, are increasingly vulnerable to machine learning-based attacks. To address this challenge, we introduce a method for generating adversarial handwritten Arabic CAPTCHAs that remain user-friendly yet difficult for machines to solve. Our approach involves synthesizing handwritten Arabic words using a simulation technique, followed by the application of five adversarial perturbation techniques: Expectation Over Transformation (EOT), Scaled Gaussian Translation with Channel Shifts (SGTCS), Jacobian-based Saliency Map Attack (JSMA), Immutable Adversarial Noise (IAN), and Connectionist Temporal Classification (CTC). Evaluation results demonstrate that JSMA provides the highest level of security, with 30% of meaningless word CAPTCHAs remaining completely unrecognized by automated systems falling to 6.66% for meaningful words. From a usability perspective, JSMA also achieves the highest accuracy rates, with 75.6% for meaningless words and 90.6% for meaningful words. Our work presents an effective strategy for enhancing the security of Arabic websites and online interfaces against bot attacks, contributing to the advancement of CAPTCHA systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: User Authentication and Security Systems · Advanced Malware Detection Techniques · Biometric Identification and Security
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, College of Computer, Qassim University, Buraydah 51452, Saudi Arabia
Funding
Deanship of Graduate Studies and Scientific Research at Qassim University
References (51)
Show source-provided references
1. Shivani, A., and Challa, R. (2020, January 30). CAPTCHA: A Systematic Review. Proceedings of the 2020 IEEE International Conference on Advent Trends in Multidisciplinary Research and Innovation (ICATMRI), Buldhana, India.10.1109/ICATMRI51801.2020.9398494 ↗
2. Bursztein, E., Martin, M., and Mitchell, J. (2011, January 17–21). Text-based CAPTCHA Strengths and Weaknesses. Proceedings of the 18th ACM Conference on Computer and Communications Security, Chicago, IL, USA.10.1145/2046707.2046724 ↗
3. A Survey of Current Research on CAPTCHA10.5121/ijcses.2016.7301 ↗
4. Batch-transformer for scene text image super-resolution10.1007/s00371-024-03598-7 ↗
5. Generative adversarial networks for handwriting image generation: A review10.1007/s00371-024-03534-9 ↗
6. Aldosari, M.H., and Al-Daraiseh, A.A. (2016, January 5–7). Strong multilingual CAPTCHA based on handwritten characters. Proceedings of the 2016 7th International Conference on Information and Communication Systems (ICICS), Irbid, Jordan.10.1109/IACS.2016.7476118 ↗
7. Cyber Security Using Arabic CAPTCHA Scheme
8. Adversarial CAPTCHAs10.1109/TCYB.2021.3071395 ↗
9. Bursztein, E., Aigrain, J., Moscicki, A., and Mitchell, J. (2014, January 19). The end is nigh: Generic solving of text-based CAPTCHAs. Proceedings of the Workshop on Offensive Technologies, San Diego, CA, USA.
10. Extended Research on the Security of Visual Reasoning CAPTCHA
11. Panda, S. (2022). Recognizing CAPTCHA using Neural Networks. Int. J. Sci. Res. Eng. Manag., 10.10.55041/IJSREM16419 ↗
12. Wang, H., Zheng, F., Chen, Z., Lu, Y., Gao, J., and Wei, R. (2018, January 15–20). A CAPTCHA Design Based on Visual Reasoning. Proceedings of the 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Calgary, AB, Canada.10.1109/ICASSP.2018.8461764 ↗
13. Hossen, M.I., and Hei, X.S. (2022, January 6–10). aaeCAPTCHA: The Design and Implementation of Audio Adversarial CAPTCHA. Proceedings of the 2022 IEEE 7th European Symposium on Security and Privacy, Genoa, Italy.10.1109/EuroSP53844.2022.00034 ↗
14. Conti, M., Guarisco, C., and Spolaor, R. (2015, January 2–5). CAPTCHaStar! A Novel CAPTCHA Based on Interactive Shape Discovery. Proceedings of the International Conference on Applied Cryptography and Network Security, New York, NY, USA.10.1007/978-3-319-39555-5_33 ↗
15. Truong, H.D., Turner, C.F., and Zou, C.C. (2011, January 5–9). iCAPTCHA: The Next Generation of CAPTCHA Designed to Defend against 3rd Party Human Attacks. Proceedings of the 2011 IEEE International Conference on Communications (ICC), Kyoto, Japan.10.1109/icc.2011.5963009 ↗
16. A Review on Evolution of various CAPTCHA in the field of Web Security10.22214/ijraset.2022.40190 ↗
17. Usuzaki, S., Aburada, K., Yamaba, H., Katayama, T., Mukunoki, M., Park, M., and Okazaki, N. (2018, January 5–8). Interactive Video CAPTCHA for Better Resistance to Automated Attack. Proceedings of the 2018 Eleventh International Conference on Mobile Computing and Ubiquitous Network (ICMU), Auckland, New Zealand.10.23919/ICMU.2018.8653624 ↗
18. A Proposed Methodology for Detecting Human Attacks on Text-based CAPTCHAs
19. Rusu, A., and Govindaraju, V. (2004, January 26–29). Handwritten CAPTCHA: Using the difference in the abilities of humans and machines in reading handwritten words. Proceedings of the 9th International Workshop on Frontiers in Handwriting Recognition (IWFHR-9 2004), Tokyo, Japan.
20. Generation and use of handwritten CAPTCHAs10.1007/s10032-009-0102-z ↗
21. Govindaraju, V., and Thomas, A. (2010). Enhancing Cyber Security through the Use of Synthetic Handwritten Captchas, Computer Science Department 226 Bell Hall.
22. Synthetic handwritten CAPTCHAs10.1016/j.patcog.2008.12.018 ↗
23. Rusu, A., Mislich, S., Missik, L., and Schenker, B. (2013, January 16–18). A Multilingual Handwriting Approach to CAPTCHA. Proceedings of the 2013 17th International Conference on Information Visualisation, London, UK.10.1109/IV.2013.26 ↗
24. On the challenges that handwritten text images pose to computers and new practical applications. Document Recognition and Retrieval XII
25. Rusu, A., and Govindaraju, V. (2005). Visual CAPTCHA with handwritten image analysis. International Workshop on Human Interactive Proofs, Springer.10.1007/11427896_3 ↗
26. Rusu, A., and Govindaraju, V. (September, January 31). A human interactive proof algorithm using handwriting recognition. Proceedings of the Eighth International Conference on Document Analysis and Recognition (ICDAR’05), Seoul, Korea.
27. Achint, T., and Venu, G. (2008, January 19–21). Generation and performance evaluation of synthetic handwritten captchas. Proceedings of the First International Conference on Frontiers in Handwriting Recognition, ICFHR, Montreal, QC, USA.
28. Random Handwritten CAPTCHA: Web Security with a Difference
29. Rusu, A.I., Docimo, R., and Rusu, A. (2010, January 23–24). Leveraging Cognitive Factors in Securing WWW with CAPTCHA. Proceedings of the USENIX Conference on Web Application Development, Boston, MA, USA.
30. Rusu, A., and Docimo, R. (2009, January 15–17). Securing the web using human perception and visual object interpretation. Proceedings of the 2009 13th International Conference Information Visualisation, Barcelona, Spain.10.1109/IV.2009.48 ↗
31. Attack-filtered interactive arabic CAPTCHAs
32. Lajmi, H., Idoudi, F., Njah, H., Kammoun, H.M., and Njah, I. (2024, January 18–20). Strengthening Applications’ Security with Handwritten Arabic Calligraphy Captcha. Proceedings of the 2024 IEEE 8th Forum on Research and Technologies for Society and Industry Innovation (RTSI), Milano, Italy.10.1109/RTSI61910.2024.10761297 ↗
33. Usability and Security of Arabic Text-based CAPTCHA Using Visual Cryptography
34. Segmentation-validation based handwritten Arabic CAPTCHA generation10.1016/j.cose.2020.101829 ↗
35. Alsuhibany, S.A., and Parvez, M.T. (2016, January 23–26). Secure Arabic Handwritten CAPTCHA Generation Using OCR Operations. Proceedings of the 2016 15th International Conference on Frontiers in Handwriting Recognition (ICFHR), Shenzhen, China.10.1109/ICFHR.2016.0035 ↗
36. Generating realistic Arabic handwriting dataset
37. Synthetic Arabic handwritten CAPTCHA
38. Alkhodidi, T., Aljoudi, L., Fallatah, A., Bashy, A., Ali, N., Alqahtani, N., Almajnooni, N., Allhabi, A., Albarakati, T., and Alafif, T.K. (2021, January 17–18). GEAC: Generating and Evaluating Handwritten Arabic Characters Using Generative Adversarial Networks. Proceedings of the 2021 International Conference on Computational Intelligence and Knowledge Economy (ICCIKE), Dubai, United Arab Emirates.10.1109/ICCIKE51210.2021.9410746 ↗
39. Educational and Cybersecurity Applications of an Arabic CAPTCHA Gamification System
40. Kim, T., Fang, W., Khan, M.K., Arnett, K.P., Kang, H., and Ślęzak, D. (2010). Using Arabic CAPTCHA for Cyber Security. Security Technology, Disaster Recovery and Business Continuity, Springer. Communications in Computer and Information Science, Vol. 122.10.1007/978-3-642-17610-4 ↗
41. Goodfellow, I.J., Shlens, J., and Szegedy, C. (2014). Explaining and Harnessing Adversarial Examples. arXiv.
42. Generative Adversarial Nets
43. Terada, T., Nguyen, V.N.K., Nishigaki, M., and Ohki, T. (2022, January 13–15). Improving Robustness and Visibility of Adversarial CAPTCHA Using Low-Frequency Perturbation. Proceedings of the International Conference on Advanced Information Networking and Applications, Sydney, Australia.10.1007/978-3-030-99587-4_50 ↗
44. No Bot Expects the DeepCAPTCHA! Introducing Immutable Adversarial Examples, With Applications to CAPTCHA Generation10.1109/TIFS.2017.2718479 ↗
45. Robust CAPTCHAs Towards Malicious OCR10.1109/TMM.2020.3013376 ↗
46. Augment CAPTCHA Security Using Adversarial Examples With Neural Style Transfer10.1109/ACCESS.2023.3298442 ↗
47. Improving the Security of Audio CAPTCHAs With Adversarial Examples10.1109/TDSC.2023.3236367 ↗
48. CAPTCHA Image Generation Systems Using Generative Adversarial Networks
49. Alsuhibany, S.A. (2023). A Survey on Adversarial Perturbations and Attacks on CAPTCHAs. Appl. Sci., 13.10.3390/app13074602 ↗
50. Google Cloud (2024, September 01). Cloud Vision API. Available online: https://cloud.google.com/vision.
51. Ghaddy (2024). Ghaddy/Adversarial: Enhancing Security of Online Interfaces: Adversarial Handwritten Arabic CAPTCHA Generation (v1.0.0). Zenodo.