Biometrics Authentication: Issues and Solutions

Abstract

Biometrics always uses the genetic characteristics of a human being to verify a person. Since this approach is popular and not the standard approach of validating a particular person likes, password, etc., it has successfully triggered much of attention. Biometrics is usually dependent on distinguishing signs characteristics of a person by face, iris, voice signature, and this pattern to ensure that each person has a unique and accurate pattern. Biometrics help the person’s identity remains remembered and not be duplicated or exchanged in comparison with the standard approach of key and password. Because of the high level of security and ease of use, biometric technology is preferred over the standard approach. These systems, which authenticate a person using fingerprint, iris, face, etc. are available in the market and are used in sensitive systems such as ATMs and access controls. Indeed, attackers keep on inventing new techniques to break through the validation process of biometric systems. They have recently used anonymous fake biometric signs for self-validation and have successfully hacked through these systems. For instance, hackers have used a photo of their targeted person and tried to deceive the authentication security systems. In this research, we propose a new validation technique that provides a higher level of security compared to traditional methods. This new technique focuses on using the human body as a source for multiple factors authentication solutions to hinder threats of system breakthroughs. We used multiple datasets to test our contribution and the acquired outcomes were high accuracy. We succeeded to recognize real biometric traits from fake ones which added value to our contribution. Besides, we conducted case studies applications in security systems based on our research. The results of these applications led us to innovations in these security systems and awarded patents due to the added enhancements and novelty.

Date of Award	2020
Original language	American English
Awarding Institution	HBKU College of Science and Engineering