Enhancing Authentication in Online Distant Exams: A Proposed Method Utilizing Face and Voice Recognition

Authors

Ahmed T. Resha, Information System Department, Faculty of Computers and Information System, Fayoum University, Fayoum, Egypt\\ Higher Institute of Marketing, Commerce & Information Systems, New Cairo, EgyptFollow
Mohammed A. Azize A. Razek, Faculty of Science, Computer Science Department, Al-Azhar University, Cairo, EgyptFollow
Eslam E. El maghraby, Information System Department, Faculty of Computers and Information System, Fayoum University, Fayoum, EgyptFollow

Abstract

Due to COVID-19 pandemic, face-to-face teaching has been replaced by online education to reduce the risks of spreading the Coronavirus. Online examination is an important asset in the context of online learning to assess students, but observing students during testing and ensuring that they do not engage in misbehavior remains a major issue. Human observation is one of the most common methods when conducting an exam to ensure that students do not perform any unexpected behaviors, by entering the student in a laboratory or hall at the university and observing him throughout the exam period visually and soundly. However, this method is costly and labor-intensive. In this paper, a system is created that monitors students during an online test automatically based on face recognition and voice recognition using a machine learning algorithm. The camera on the students computer will be used to track the students facial movements, pupils, and lip movements, monitoring the students behavior throughout the test, and stopping any unexpected behavior. In this system, there are two parts: facial recognition and unexpected behavior detection. The face was recognized with an accuracy of 98.3%, and unexpected behavior was detected with an accuracy of 97.6%. There is also an opportunity to increase accuracy by improving the quality of the images in the dataset.

Recommended Citation

T. Resha, Ahmed; A. Azize A. Razek, Mohammed; and E. El maghraby, Eslam (2023) "Enhancing Authentication in Online Distant Exams: A Proposed Method Utilizing Face and Voice Recognition," Information Sciences Letters: Vol. 12 : Iss. 6 , PP -.
Available at: https://digitalcommons.aaru.edu.jo/isl/vol12/iss6/34