Title :
Gestalt geometric CAPTCHA
Author :
Nuttanont Hongwarittorrn;Suttikiat Meelap
Author_Institution :
Department of Computer Science, Faculty of Science and Technology, Thammasat University, Pathumthani, Thailand
Abstract :
This research investigated a new Image-Based CAPTCHA called Gestalt Geometric CAPTCHA, which does not require the use of a database of images, and is based on the Gestalt Theory of human recognition. The aim was to develop a type of CAPTCHA that is easier for human users and harder for bots. We experimentally tested the use and effectiveness of Gestalt Geometric CAPTCHA in terms of time for completion, authentication pass rate, and user satisfaction, in comparison with reCAPTCHA, Ironclad CAPTCHA, and ShapeCAPTCHA. We also tested our novel CAPTCHA for robustness against two shape detection and classification programs, ShapeChecker and Shape-detect. The results were promising, but some issues remain for future improvement.
Keywords :
"CAPTCHAs","Standards","Robustness","Psychology","Distortion","Image recognition","Semantics"
Conference_Titel :
Advanced Computer Science and Information Systems (ICACSIS), 2015 International Conference on
DOI :
10.1109/ICACSIS.2015.7415185