Downloads: 4 | Views: 146 | Weekly Hits: ⮙1 | Monthly Hits: ⮙4

Masters Thesis | Computer Science | India | Volume 11 Issue 3, March 2022 | Rating: 5.2 / 10

Online Proctoring System to Avoid Defrauding

K. Pranay Krishna | Dr. G. Arun Sampaul Thomas | B. Manasa

Abstract: Online remote control has thrived during these Covid19 pandemic conditions. All schools and educational institutions have been closed and students are now completing their studies through online applications. However, there are significant difficulties in carrying out the assessments. Some schools have switched to an assignment format, where students can simply copy and paste their answers from the Internet. If our way of life becomes the new normal, a solution is needed. It is difficult to allow students to take the test at home over the Internet while being supervised by a supervisor for the duration of the exam. Due to the effort, it is not possible to continue to apply this method on a large scale. To solve this problem, we developed an AI in Python that can observe students using their laptops or the system's webcam and microphone, allowing supervisors to monitor multiple students at the same time. Our AI has four vision-based capabilities tied together so they can work together. Gaze tracking, mouth opening or closing, person counting, and cell phone detection are four features. In addition, the voiceover from the microphone is captured, converted to text, and compared with the test text to show the amount of frequently reported words by each student taking the individual test.

Keywords: Online Proctoring, face recognition, Tabs switching, Head pose

Edition: Volume 11 Issue 3, March 2022,

Pages: 1430 - 1432