Abstract—CAPTCHA is almost a standard security
technology, and has found widespread application in
commercial websites. Image recognition CAPTCHAs faces
many potential problems which have not been fully studied. It is
difficult for a small site to acquire a large dictionary of images
which an attacker does not have access to and without a means
of automatically acquiring new labeled images, an image based
challenge does not usually meet the definition of a CAPTCHA.
They are either unusable or prone to attacks. In this paper, we
propose the Character Image Semantic (CIS) based CAPTCHA
that will combine both Image and Text into one CAPTCHA
which will strengthen it and does not make it human solvable.
Benchmarks will be discussed after conducting both Usability
and Security tests with a clear set of results that show the
accuracy of the proposed solution compared to other existing
ones.
Index Terms—CAPTCHA, security, image recognition,
semantics.
Aziz Barbar is with American University of Science & Technology,
Beirut, Lebanon (e-mail: abarbar@aust.edu.lb).
Anis Ismail is with University Institute of Technology, Lebanese
University, Sidon, Lebanon (e-mail: anismaiil@ul.edu.lb).
[PDF]
Cite: Aziz Barbar and Anis Ismail, "Character Image Semantic-Based CAPTCHA," International Journal of Future Computer and Communication vol. 4, no. 3, pp. 211-215, 2015.
