aRaspberry Pi-Based Face Recognition by Using Eigenface Algorithm to Detect Cheating on Exams

Abstract

Cheating is an act of lying, trickery or coercion carried out to create an unfair advantage, often
at the expense of someone else. A common area for frequent cheating is in examinations.
Currently, supervision in examinations has been improved, but it is still difficult for supervisors
to find out who actually cheated the exam. The identity fraud technology used in exams is getting
more advanced. Jockeying is a common method often used for cheating on exams. Therefore,
how to correctly identify the fake identity has become a difficult problem. Based on several
studies that have been conducted, we developed face recognition as a test fraud detector. Where
to design the raspberry pi and picamera as tools for taking pictures, then the data is processed
and its status determined using the eigenface algorithm which will then send the data to the
database server. So it can be proven that when a participant is detected cheating, the system will
automatically lock the exam process.