Novel Authentication System for Personal and Domestic Network Systems Using Image Feature Comparison and Digital Signatures

Abstract

With the advent of the digital age, there has been an increased need for the robustness and the versatility of authentication systems. Many techniques have been developed since the last few decades in digital security to secure and encrypt data. As much as these techniques are versatile and robust, authentication systems remain the weakest link in any given cyber-physical system. Human intervention does not necessarily make a system as robust as it ought to be, more so often, the chances are it might result in adverse and also opposite of the desired effect. In order for a system to be robust and secure, all forms of precaution must be in place for it to function securely. Authentication systems, since its inception are designed largely in order to minimize human errors and to securely authenticate and verify an individual’s identity in question. Traditional authentication systems are either not efficient and or can be easily bypassed. This could be largely rectified or reduced using modern authentication systems like bio-metrics, iris detection systems etc. However, these systems are not easy to setup and configure, being highly sophisticated and require trained personnel to maintain. Through this paper, we propose a novel authentication system which is relatively inexpensive, easily made mobile and secure to setup. The system targets domestic networks and aims to secure personal and small scale systems. This novel authentication scheme proposed operates based on image feature comparison and digital signatures. The image comparison is used for the authentication process itself while a digital signature is used to provide non-repudiation. Through this novel approach, we aim to demonstrate and highlight the efficiency of the proposed system to the other existing authentication schemes used today. This approach not only provides an easy way to authenticate, it provides a way to determine if the person has indeed the authorization to access the system.

https://github.com/hrishikesh9409/ImageAuthentication/implementation.