Design Paper for Proctored Exam System Using Video Authentication

Abstract

In the past two years, due to the COVID-19 situation, there has been a need for various online examinations which were conducted by various universities across the world.. Thus, the use of machine learning tools and various other educational tools that can be operated remotely has been used continuously to increase reach among people who don't have access to organisations. Exams are a more important factor in the educational system, an online learning process in the student's career. Since by various methods have been invented to provide efficient, comfortable online exam proctoring to reduce malpractice in the examination by the student. Various malpractices can be done from the student side which should be halted. For that various delivery tools as well as systems have been created to conduct the exams in online mode from any of the places. Those tools will help to reduce time consumption. This paper introduces a new approach for the "Online Exam Proctoring system” using the Haar cascade algorithm. Thus, by this system there is not only assurance for the identification of a test-taker but also the overall integrity of the test there should be transparency in the system because mainly online exam security is a major concern. In this system, there is the usage of Haarcascade algorithm which will be used for the feature extraction and will give the system accuracy to enhance the exam security and to minimize the stressful restrictions.