Enhancing CAPTCHA Using Eye Movement

Abstract

CAPTCHA is a method designed to prevent automatic registration and access. It can distinguish between real human and automated program (BOT). CAPTCHA connects to a server and then starts human verification process. The user must answer a question correctly or follow an instruction in order to get access to the system. However, there are some problems occurred to users. For example, the verification text is too distorted that even real humans do not understand or the effectiveness of preventing an automated computer program is doubted. From these reasons, researchers are still developing CAPTCHA to make it better and more effective. In this research, the proposed CAPTCHA called EYE CAPTCHA is mixed between CAPTCHA and eye movement technology. EYE CAPTCHA consists of three important components. The first component is the puzzle-based CAPTCHA. The second component is the gaze detection which is based on the relation between an eye and a position on the screen. The last component of this proposed method is the eye movement for moving objects to the desired direction. From an experiment and statistical analysis, this method can enhance the performance of CAPTCHA in distinguishing between humans and bots.