Facial Recognition System to Detect Student Emotions and Cheating in Distance Learning

Abstract

Distance learning has spread nowadays on a large scale across the world, which has led to many challenges in education such as invigilation and learning coordination. These challenges have attracted the attention of many researchers aiming at providing high quality and credibility monitoring of students. Distance learning has offered an effective education alternative to traditional learning in higher education. The lecturers in universities face difficulties in understanding students’ emotions and abnormal behaviors during educational sessions and e-exams. The purpose of this study is to use computer vision algorithms and deep learning algorithms to develop a new system that supports lecturers in monitoring and managing students during online learning sessions and e-exams. To achieve the proposed objective, the system employs software methods, computer vision algorithms, and deep learning algorithms. Semi-structural interviews were also used as feedback to enhance the system. The findings showed that the system achieved high accuracy for student identification in real time, student follow-up during the online session, and cheating detection. Future work can focus on developing additional tools to assist students with special needs and speech recognition to improve the follow-up facial recognition system’s ability to detect cheating during e-exams in distance learning.

New Trends in Emotion Recognition Using Image Analysis by Neural Networks, a Systematic Review

“…There are also situations in which the proposed methods exceed the state-of-the-art performances [ 38 , 39 , 54 , 62 , 74 , 81 , 101 ]. Similarly, context-aware solutions for emotion recognition [ 47 , 49 , 50 , 98 ] or practical solutions [ 37 , 124 , 127 , 130 , 133 , 159 ] demonstrate promising results.…”

“…Further, the Xception architecture [ 125 ] abstracts the input of each layer so that in the end it obtains a compact representation of each layer from which a single value is obtained, representing the prediction. The Xception network is used in [ 54 ] for feature extraction and in [ 126 , 127 , 128 ] is used as a data segregator in a pre-trained model.…”

Learning analytics with correlation-based SAN-LSTM mechanism for formative evaluation and improved online learning

“…Data warehouses (DW) have been used in certain research to examine the ways students make decisions while utilizing the learning management system (LMS) to further decide based on their assessment [13,14]. Using online tools and resources like LMS, students obtain a degree through distant learning after passing the examination [15]. Students benefit from methods of instruction in online learning, and teachers assist students in restoring interest for studying in a variety of ways, such as by providing educational resources [16].…”

Finger-vein Secure Access Based on Lightweight Dual-attention Convolutional Neural Network for Quality Distance Education

“…(7) Additionally, identity recognition technology is essential to maintain the quality, fairness, and security of distance learning. Using identity recognition not only ensures that students participating in online courses and exams are not imposters, (8) but it also prevents unauthorized access to personal data and enables teaching platforms to offer a more personalized learning experience based on individual student progress and preferences. (9) In recent years, with the development of computer vision (CV) and deep learning technologies, and driven by the COVID-19 pandemic, the intelligent learning industry market has become increasingly active.…”