A Novel Construction of Correlation-Based Image CAPTCHA with Random Walk
CAPTCHA has been widely adopted throughout the World Wide Web to achieve network security by preventing malicious interruption or abuse of server resources. Existing text-based and image-based CAPTCHA techniques are not robust enough to resist sophisticated attacks using pattern recognition and machine learning. To overcome this challenge, we designed a new approach to construct an image-based CAPTCHA by using a random walk on image with correlated contents, which capitalizes on human knowledge on the relevance of images. The usability and robustness of the proposed scheme have been evaluated by both numerical analysis and empirical evidence. Early testing has shown it to be a promising approach to enhancing and replacing the existing Web CAPTCHA techniques when fighting against bots.
KeywordsCAPTCHA Image correlation Random walk
This material is based upon work supported by the China NSF grant No. 61472189, the CERNET Innovation Project No. NGII20160601, and the Innovation Projects of Beijing Engineering Research Center of Next Generation Internet and Applications.
The authors confirm that an ethic approval for this particular type of study is not required in accordance with the policy of the involved institutes.
- 1.Yan, J., & Ahmad, A. S. E. (2011). Captcha robustness: a security engineering perspective. Computer, 44(2), 54–60.Google Scholar
- 2.Bursztein, E., Martin, M., & Mitchell, J. (2011). Text-based CAPTCHA strengths and weaknesses. ACM Conference on Computer and Communications Security, CCS 2011, Chicago, Illinois, Usa, October (pp. 125–138). DBLP.Google Scholar
- 3.Datta, R., Li, J., & Wang, J. Z. (2009). Exploiting the human–machine gap in image recognition for designing captchas. IEEE Transactions on Information Forensics & Security, 4(3), 504–518.Google Scholar
- 4.He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep Residual Learning for Image Recognition. Computer Vision and Pattern Recognition (pp. 770–778). IEEE.Google Scholar
- 5.Lopezpaz, D., Hennig, P., & Schölkopf, B. (2013). The randomized dependence coefficient. Advances in Neural Information Processing Systems, 1–9.Google Scholar
- 6.Fouss, F., Pirotte, A., Renders, J. M., & Saerens, M. (2007). Random-walk computation of similarities between nodes of a graph with application to collaborative recommendation. IEEE Transactions on Knowledge & Data Engineering, 19(3), 355–369.Google Scholar
- 7.Chew, M., Tygar, J.D. (2004) Image recognition captchas. In: Zhang, K., Zheng, Y. (eds.) ISC 2004. LNCS, vol. 3225, pp. 268–279. Springer, Heidelberg.Google Scholar
- 8.Rahman, R. U., Tomar, D. S., & Das, S. (2012). Dynamic Image Based CAPTCHA. International Conference on Communication Systems and Network Technologies (pp. 90–94). IEEE.Google Scholar
- 9.Sivakorn, S., Polakis, I., & Keromytis, A. D. (2016). I am Robot: (Deep) Learning to Break Semantic Image CAPTCHAs. IEEE European Symposium on Security and Privacy (pp. 388–403). IEEE.Google Scholar
- 10.Moawad, I. F., & Aref, M. (2013). Semantic graph reduction approach for abstractive Text Summarization. Seventh International Conference on Computer Engineering & Systems (pp. 132–138). IEEE.Google Scholar
- 11.Lee, J., Cho, M., & Lee, K. M. (2010). A Graph Matching Algorithm Using Data-Driven Markov Chain Monte Carlo Sampling. International Conference on Pattern Recognition, ICPR 2010, Istanbul, Turkey, 23–26 August (pp. 2816–2819). DBLP.Google Scholar