• DocumentCode
    1058986
  • Title

    CAPTCHAs: Humans vs. Bots

  • Author

    Kolupaev, Aleksey ; Ogijenko, Juriy

  • Volume
    6
  • Issue
    1
  • fYear
    2008
  • Firstpage
    68
  • Lastpage
    70
  • Abstract
    A completely automated public turing test to tell computers and humans apart (CAPT-CHA) offers a way for Web service providers to make some conclusions about whether a "user" is human or robot. Process of CAPTCHA recognition is a combination of efforts, approaches, and software that attempts to increase accuracy to an acceptable level. Of course, it\´s hard to define this level, but we\´ve found that it often starts from 5 percent accuracy. We\´ve covered the most popular CAPTCHA approaches here but not all of them. Most users know how to use the more commonly used graphical CAPTCHAs, so designers must have serious reasons to migrate to other, more difficult-to-break, approaches. After all, CAPTCHAs aim to determine legitimate users, not alienate them.
  • Keywords
    Web services; humanoid robots; optical character recognition; security of data; CAPTCHA recognition; Web service provider; completely automated public turing test to tell computers and humans apart; Character recognition; Computer crime; Computer hacking; Costs; Humans; Image recognition; Optical character recognition software; Protection; Security; Web services; CAPTCHA; Completely Automated Public Turing test to tell Computers and Humans Apart; Crypto Corner; cryptography; optional character recognition;
  • fLanguage
    English
  • Journal_Title
    Security & Privacy, IEEE
  • Publisher
    ieee
  • ISSN
    1540-7993
  • Type

    jour

  • DOI
    10.1109/MSP.2008.6
  • Filename
    4446701