Gaze Gesture Authentication System Against Shoulder Surfing Attacks
Manish Pillai1, Srivenkatesh Nair 2, Sujith Kurup3, Abhinav Menon4, and Prof. K.S. Charumathi5
1BE Student, Department of Computer Engineering, Pillai College of Engineering, Navi Mumbai - 410206
2BE Student, Department of Computer Engineering, Pillai College of Engineering, Navi Mumbai - 410206
3BE Student, Department of Computer Engineering, Pillai College of Engineering, Navi Mumbai - 410206
4BE Student, Department of Computer Engineering, Pillai College of Engineering, Navi Mumbai - 410206
5Assistant Professor, Department of Computer Engineering, Pillai College of Engineering, Navi Mumbai - 410206
---------------------------------------------------------------***------------------------------------------------------------------
Abstract - Shoulder surfing is the term used to describe one person observing another person’s computer or mobile device screen and keyboard to obtain sensitive information. Direct observation can be done by simply looking over someone’s shoulder – hence shoulder surfing – or using binoculars, video cameras, and other optical devices. Shoulder surfing usually has the goal of viewing and stealing sensitive information such as username and password combinations that can be used to enter a user's account later. Credit card numbers, PIN, and sensitive personal information used in response to security questions are also targeted. So, the proposed Two-factor authentication would secure the system from the Shoulder Surfing Attack which is disparate from the traditional existing system where username and password are required for authentication. In the proposed system, the first phase is username & password validation just like a login page. The information for authentication is provided by the user while registering to the page. In the next phase, Gaze Gesture Authentication is used through Real time eye tracking using OpenCV and Dlib. Through multiple evaluations, we discuss how the authentication accuracy varies with respect to transition speed of numbers and user’s blink.
Keywords: Shoulder Surfing, Gaze Gesture, Authentication, Security, Eye Tracking, Patterns.