DEVELOPMENT OF A FACIAL RECOGNITION-BASED ALERT SECURITY SYSTEM

Student: Imisioluwa Emmanuel Osunnuyi
Supervisor: Mr Osungbemi Adeleke
HOD: Mrs Latifat Odeniyi
Department of Computer Engineering
Engineering
Olabisi Onabanjo University, Ago-Iwoye, Ogun State

Abstract

With the rapid advancements in biometrics technology, I propose a facial recognition-based home security system that uses a Raspberry Pi and includes a telegram alert functionality. This system enhances security by recognizing individuals through facial recognition and triggering appropriate actions based on identity verification. The Raspberry Pi serves as the core control unit, utilizing a camera module to capture facial images and a machine learning model is trained to recognize faces and compare them against a database of authorized individuals. Upon recognizing a person, the system either grants or denies access, activating or deactivating security measures according to the response of the administrator. Additionally, the system sends alerts to notify unauthorized access attempts, providing details such as the intruder's image, date, and time. This real-time notification ensures a quick response to potential security breaches. The system offers benefits such as real-time monitoring, enhanced accuracy in identifying authorized individuals, and immediate alerts of security issues. It provides a reliable and robust home security solution by leveraging IoT and facial recognition technology. Through extensive testing under various scenarios, the system demonstrated high accuracy in facial recognition, swift response times, and reliable email notifications, proving its effectiveness in improving home security.

Full-Text Access Notice

In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication. During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.

Request authorisation from the author