MULTI-FACTOR AUTHENTICATION SYSTEM BASED ON TEMPLATE

Abstract

User authentication is an integral part of a secure system. Traditional authentication methods have been proven to be vulnerable to various types of security attacks. Artificial intelligence is used to crack plain text passwords, and even CAPTCHAs are removed in a few tries. Using a picture password as an alternative to plain text passwords to authenticate users can be an effective strategy. However, they have been proven to be susceptible to shoulder attacks when surfing. Advanced authentication systems such as biometrics are secure, but they require additional infrastructure to be implemented effectively.      This document proposes a new template-based multi-factor authentication scheme that uses a combination of text and images to identify legitimate users. The proposed system was mathematically analyzed and found to provide much more space for passwords than plain text passwords. This makes the proposed system safe from brute force and other dictionary attacks. Plus, using text along with images also reduces the risk of shoulder surfing.