Development of a neural Network Face Biometrics System for Impersonation Detection in an Examination

Abstract

Impersonation is one of the worst and most unsettling types of examination malpractice; it involves a collaboration between students and the examiners, lecturers, and other students to allow an unregistered (but brilliant) student to take the exam in place of a registered one. Obviously, this has a disastrous impact on the entire educational system. The construction of a software system to identify impersonators while being examined is the main focus of the paper. The system is built with Deep Convolutional Neutral Network (DCNN) biometric technology, which is based on identifying a person’s face traits, and Model View Controller (MVC) image processing approach. MySQL was used as the database, and Python was used to implement it. Data was gathered through primary and secondary sources, and the face biometric dataset will be purchased online. The research resulted in the development of an impersonation detection system during examination for students at Federal Polytechnic, Auchi, in order to establish an environment conducive to free and fair examination.