.png){kind=icon}
.png){kind=icon}
Return
Automated smart artificial intelligence-based proctoring system using deep learning
.png){kind=icon}
PRE
AI
DOI：10.1007/s00500-023-08696-7.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Since COVID-19, there have been significant advancements made in the area of online teaching and learning. To provide their pupils with more resources, academic institutions are going digital. Students now have more options for learning at their speed and developing their skills. There has been a shift in favor of online tests for evaluations. AI-assisted proctoring solutions are in great demand as online proctoring services grow in popularity. We provide a method for doing away with the need for a human proctor to be present during the test by creating a multi-modal system. To get footage, we used a camera and active window capture. To infer the test taker's emotions, his face is recognized. To establish his head position, his feature points are calculated. The surroundings of the examinee can be picked up on, such as a phone, a book, or the presence of another person. Additionally, our system also keeps track of the examinee's mouth opening and face spoofing. An intelligent rule-based inference system that can determine whether or not there was examination fraud is produced by the combination of these models.
Keywords:
Online examination proctoring
Face detection and emotion recognition
Head pose estimation
Deep learning
Face spoofing detection
Journal
IF:
2.5
Papers:
1.0W
Citations:
2.1W