论文信息 - Investigating CAPTCHAs Based on Visual Phenomena

Investigating CAPTCHAs Based on Visual Phenomena

We propose and evaluate several novel types of CAPTCHAs (test to tell computers and humans apart) that exploit characteristics of the human visual system. Perceptions caused by the effect of lightness constancy or grouping phenomena due to transparent motion are hard to emulate on computers and may thus provide novel authentication mechanisms.

Christian Bauckhage | Thomas Franke | Anja Naumann

[1] David Regan. Detection and discrimination of motion-defined and luminance-defined two-dimensional form , 1994 .

[2] K. Herzog,et al. What is visual intelligence , 1997 .

[3] Patrice Y. Simard,et al. Using Machine Learning to Break Visual Human Interaction Proofs (HIPs) , 2004, NIPS.

[4] Aleksey Kolupaev,et al. CAPTCHAs: Humans vs. Bots , 2008, IEEE Security & Privacy.

[5] James Ze Wang,et al. IMAGINATION: a robust image-based CAPTCHA generation system , 2005, ACM Multimedia.

[6] Jon Howell,et al. Asirra: a CAPTCHA that exploits interest-aligned manual image categorization , 2007, CCS '07.