Design and Implementation of Embedded Biometric-Based Access Control System with Electronic Lock using Raspberry Pi

Abstract

This paper presents the design and implementation of an improved access control system based on biometric recognition, utilising Raspberry Pi technology. The proposed system aims to enhance the security of the existing electronic lock-based system at the SGRE-Lab of University Tahri Mohammed of Bechar in Algeria. The proposed system employs multimodal biometrics, integrating facial recognition and speaker verification for personal identification. Following initial verification by the electronic lock, the system captures the user's face through a camera to perform facial recognition. In cases where the user's identity is uncertain, a voice recognition module prompts the user to say a secret word, confirming their identity through the microphone. The combination of these two biometric techniques ensures access is granted, and an access log is recorded, with an accompanying notification sent to the administrator via SMS. As technical contribution, this paper presents the design and implementation of an embedded biometric-based access control system using Raspberry Pi, which includes the integration with an electronic lock and digicode, in the other hand, a second innovation contribution by combining biometric-based authentication with Raspberry Pi technology, this paper introduces an innovative approach to access control systems that provides a more secure and reliable means of access control than traditional methods based on keys or passwords. An overview of the proposed system's architecture is provided, its operation modes, and necessary hardware and software requirements. The promising obtained results of demonstrations show a notable improvement in security levels, characterized by reduction of false acceptances, however, the paper acknowledges that users unfamiliar with the biometric system may face challenges, potentially leading to false rejections. Future work should focus on mitigating these challenges and addressing user familiarity issues.