On random field CAPTCHA generation

  • Author(s) / Creator(s)
  • Herein, we propose generating CAPTCHAs through random field simulation and give a novel, effective and efficient algorithm to do so. Indeed, we demonstrate that sufficient information about word tests for easy human recognition is contained in the site marginal probabilities and the site-to-nearby-site covariances and these quantities can be embedded into KNW conditional proba- bilities, designed for effective simulation. The CAPTCHAs are then partial random realizations of the random CAPTCHA word: we start with an initial random field (e.g., randomly scattered letter pieces) and use Gibbs resampling to re-simulate portions of the field repeatedly using the KNW conditional probabilities until the word becomes human-readable. The residual randomness from the initial random field together with the random implementation of the CAPTCHA word provide significant resistance to attack. This results in a CAPTCHA which is unrecognizable to modern OCR but is recognized about 95% of the time in a human readability study.

  • Date created
    2013
  • Subjects / Keywords
  • Type of Item
    Article (Published)
  • DOI
    https://doi.org/10.7939/R3028PD1Q
  • License
    © 2013 IEEE Transactions on Image Processing. This version of this article is open access and can be downloaded and shared. The original author(s) and source must be cited.
  • Language
  • Citation for previous publication
    • M. A. Kouritzin, F. Newton and B. Wu. (2013). " On random field CAPTCHA generation '', IEEE Transactions on Image Processing 22(4), pp. 1656-1666.