Abstract:
In recent years, online examinations have been increasingly utilized to assess
students' abilities to teach knowledge, particularly during the COVID-19 outbreak.
However, due to the lack of face-to-face connection, proctoring online tests is
difficult. Furthermore, past study has demonstrated that online examinations are
vulnerable to nu- serous duping techniques, which can compromise their legitimacy.
Specifically, this project detects suspected student by user verification, Head-Pose
Analysis, Gaze Estimation, Person Counting, Active Window detection, Mobile
Phone Detection, Multiple Monitor Detection.
Massive open online courses (M O O Cs) and other kinds of distance learning are
growing in popularity and reach. The ability to successfully proctor remote online
tests is a key limiting issue in this next stage of education's scalability. Human
proctoring is currently the most frequent method of assessment, which involves
either forcing test takers to visit an examination center or watching them visually and
audibly throughout exams via a digital camera.
