ABSTRACT
The Asirra CAPTCHA [EDHS2007], proposed at ACM CCS 2007, relies on the problem of distinguishing images of cats and dogs (a task that humans are very good at). The security of Asirra is based on the presumed difficulty of classifying these images automatically.
In this paper, we describe a classifier which is 82.7% accurate in telling apart the images of cats and dogs used in Asirra. This classifier is a combination of support-vector machine classifiers trained on color and texture features extracted from images. Our classifier allows us to solve a 12-image Asirra challenge automatically with probability 10.3%. This probability of success is significantly higher than the estimate of 0.2% given in [EDHS2007] for machine vision attacks. Our results suggest caution against deploying Asirra without safeguards.
We also investigate the impact of our attacks on the partial credit and token bucket algorithms proposed in [EDHS2007]. The partial credit algorithm weakens Asirra considerably and we recommend against its use. The token bucket algorithm helps mitigate the impact of our attacks and allows Asirra to be deployed in a way that maintains an appealing balance between usability and security. One contribution of our work is to inform the choice of safeguard parameters in Asirra deployments.
- ASR Asirra: A Human Interactive Proof. On the Web at http://research.microsoft.com/asirra/Google Scholar
- BotBarrier.com. On the web at http://www.botbarrier.com/Google Scholar
- Chih-Chung Chang and Chih-Jen Lin. LIBSVM : a library for support vector machines, 2001. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm Google ScholarDigital Library
- . Chow, P. Golle, M. Jakobsson, X. Wang and L. Wang. Making CAPTCHAs Clickable. In Proc. of HotMobile 2008. Google ScholarDigital Library
- . Cortes and V. Vapnik. Support-vector network. Machine Learning 20, 273--297, 1995. Google ScholarDigital Library
- . Douceur and J. Elson. Private communication.Google Scholar
- . Elson, J. Douceur, J. Howell and J. Saul. Asirra: a CAPTCHA that exploits interest-aligned manual image categorization. In Proc. of ACM CCS 2007, pp. 366--374. Google ScholarDigital Library
- . Golle and D. Wagner. Cryptanalysis of a Cognitive Authentication Scheme. In Proc. of the 2007 IEEE Symposium on Security and Privacy, pp.66--70. IEEE Computer Society Google ScholarDigital Library
- Google CAPTCHA. On the web at https://www.google.com/accounts/DisplayUnlockCaptchaGoogle Scholar
- . Hastie, R. Tibshirani and J. Friedman. The Elements of Statistical Learning (Data Mining, Inference, and Prediction). Springer Series in Statistics, 2001.Google ScholarCross Ref
- . Kruizinga, N. Petkov and S.E. Grigorescu. Comparison of texture features based on Gabor filters. In Proc. of the 10th International Conference on Image Analysis and Processing (1999), pp. 142--147. Google ScholarDigital Library
- . Lopresti. Leveraging the CAPTCHA problem. In Proc. of the Second International Workshop on Human Interactive Proofs, pp. 97--110. Springer Verlag, 2005. Google ScholarDigital Library
- . Mironov and L. Zhang. Applications of SAT Solvers to Cryptanalysis of Hash Functions. In Theory and Applications of Satisfiability Testing -- SAT 2006, pp. 102--115, 2006. Google ScholarDigital Library
- . Mori and J. Malik. Recognizing objects in adversarial clutter: Breaking a visual CAPTCHA. In Proc. of the 2003 Conference on Computer Vision and Pattern Recognition, pp. 134--144. IEEE Computer Society, 2003. Google ScholarDigital Library
- SlashDot. Yahoo CAPTCHA Hacked (posted Jan 29, 2008). On the Web at http://it.slashdot.org/it/08/01/30/0037254.shtmlGoogle Scholar
- Websense Blog (posted Feb 22, 2008). Google's CAPTCHA busted in recent spammer tactics. On the web at http://securitylabs.websense.com/content/Blogs/2919.aspxGoogle Scholar
- . Yan and A. El Ahmad. A Low-cost Attack on a Microsoft CAPTCHA. To appear in Proc. of ACM CCS 2008. Google ScholarDigital Library
Index Terms
- Machine learning attacks against the Asirra CAPTCHA
Recommendations
Face recognition CAPTCHA made difficult
WWW '14 Companion: Proceedings of the 23rd International Conference on World Wide WebA CAPTCHA is a Turing test to distinguish human users from automated scripts to defend against internet adversarial attacks. As text-based CAPTCHAs (TBC) have become increasingly difficult to solve, image-based CAPTCHAs, and particularly face ...
DDIM-CAPTCHA: A Novel Drag-n-Drop Interactive Masking CAPTCHA against the Third Party Human Attacks
TAAI '13: Proceedings of the 2013 Conference on Technologies and Applications of Artificial IntelligenceA CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) is a security mechanism that can be used to distinguish between humans and machines. Most existing CAPTCHA systems are vulnerable against a so-called "third party ...
Pitfalls in CAPTCHA design and implementation: The Math CAPTCHA, a case study
We present a black-box attack against an already deployed CAPTCHA that aims to protect a free service delivered using the Internet. This CAPTCHA, referred to as ''Math CAPTCHA'' or ''QRBGS CAPTCHA'', requests the user to solve a mathematical problem in ...
Comments