Open Access Open Access  Restricted Access Subscription or Fee Access

Knowing Captchas Using Deep Learning

laasya kashimalla, Pranitha A, T SoumyaSree

Abstract


Websites can use CAPTCHA (Completely Automated Public Turing Test to Differentiate Computers from Humans) to distinguish between humans and machines and prevent automated attacks by programmers and malicious programmers. Studies on CAPTCHA recognition have the potential to uncover security flaws, advance CAPTCHA technology, and even advance handwriting and license plate recognition. However, the whole objective of the CAPTCHAs can be bypassed by utilizing the principles of deep learning and computer vision. Convolutional neural networks (CNN) enable the automatic passing of this test. A CNN is a deep learning algorithm that uses an image as input before giving distinct features in the image a value that further aids in differentiating one feature from another. Its major objective is without losing features that are crucial for obtaining an optimized forecast, to convert the photos into a form that is much easier to handle. Future work on this project will involve expanding the CAPTCHA recognition algorithm to handle noisier, bigger and CAPTCHAs with all conceivable symbols.

Full Text:

PDF

References


von Ahn L, Blum M, Hopper NJ, Langford J. Captcha: using hard ai problems for security. In: EUROCRYPT; 2003:294–311. doi: 10.1007/3–540–39200–9_18.

Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans Pattern Anal Mach Intell. 2017;39(6):1137–49. doi: 10.1109/TPAMI.2016.2577031.

Gao H, Wang W, Qi J, et al. ’The Robustness of Hollow CAPTCHAs,’ Proc. ACM Conference on Computer and Communications Security pp. 2013:1075–86.

Bursztein E, Martin M, Mitchell J. Text-based CAPTCHA Strength, and Weaknesses. In: Proceedings of the ACM conference on computer and communications security; 2011. p. 125–37.

Gao H, Yan J, Cao F, Zhang Z, Lei L, Tang M, et al. A Simple Generic Attack on Text Captchas. In: Proceedings of the network and distributed system security symposium; 2016. doi: 10.14722/ndss.2016.23154.

Chen J, Luo X, Hu J, Ye D, Gong D. An Attack on Hollow CAPTCHA Using Accurate Filling and Nonredundant Merging. IETE Tech Rev. 2018;35;sup1:106–18. doi: 10.1080/02564602.2018.1520152.

Kopp M, Nikl M, Holena M. Breaking CAPTCHAs with convolutional neural networks. ITAT. 2017 Sep:(93–9).

Yu H, Riedl MO. Automatic generation of game-based CAPTCHAs. In: Proceedings of the FDG workshop on procedural content generation 2015.


Refbacks

  • There are currently no refbacks.