A new avatar dynamic image based CAPTCHA service on cloud for Mobile devices

Use of Mobile devices and internet developed gradually into a ubiquitous network and web has taken an increasingly central role in our communication. Recently the emergence of intelligent, sophisticated attacks like DOS (Denial of Service) makes web services more vulnerable than ever. To handle the DOS attacks, CAPTCHAs are used as one of the standard security technology. CAPTCHAs generate grade tests that most humans can pass but current computer programs can't. Human users and automated scripts can be classified and CAPTCHA is one of the possible ways to classify them. Text based CAPTCHAs are used in many applications but they face a challenge due to language dependency. Dynamic image based CAPTCHAs can strength the robustness of CAPTCHAs. It improves the human success rate. The probability of a BOT hacking the system is considerably low as compared to existing text based CAPTCHA. In this paper, an image-based Avatar CAPTCHA for mobile devices has been proposed as a potential solution. The solution of CAPTCHA is that users must correctly identify visually distorted human faces embedded in a complex background without selecting any non-human Avatar face. With this proposed work, CAPTCHA service is hosted on cloud that will generate CAPTCHA on cloud, so it improves efficiency of CAPTCHA generation also reduces response time for the use to authenticate as human. The need for usability and presentation that is required for a mobile device is considered during implementation. The proposed algorithm generates a CAPTCHA that offers better accuracy and lower machine attack rates compared to existing approaches.