Smart Exam Malpractice Detection and Evidence Capture YOLO
International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)
DOI: 10.15662/ijeetr.2026.0802168 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von C. Leelavathi und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Due to the increasing popularity of the online classes and tests, the proctoring systems that are both dependable and automated are needed more to ensure that the academic honesty of the students is secured. Old-fashioned ways of invigilation cannot be applied on a large scale, and they are usually not able to notice some minor indicators of malpractice in case of remote exams. The smart monitoring system is developed on the principle of using computer vision and DL to detect cheating during exams and document evidence that could be verified. To maintain a watch on suspicious activities, the system employs the publicly available datasets and detection models that are already trained. Such models are facial landmark predictors and mobile object detection models. Video frames of a webcam stream or uploaded file are preprocessed and before processing, they are executed by extracting frames, resizing, normalizing and localizing landmarks of the faces in the video frame. Various DL object recognition models, such as YOLOv5, YOLOv8, YOLOv9 and YOLOv11, are run, and their performance in terms of detection of banned items such as cell phones is tested. Performance is determined by metrics such as Precision, Recall and mAP. According to the tests, YOLOv5 performs best in terms of the aggregate performance and has a Precision of 0.848 and mAP of 0.663. Instead, YOLOv8 and YOLOv11 are the ones with the highest Recall of 0.709. The structure is compatible with a web application based on Flask in order to enable real-time tracking, evidence recording, and automatic notifications. This increases the openness and trustworthiness of AI-based test proctoring systems
Zitieren
APA
Leelavathi, C., Navya, K., Yeshwanthi, R., & Venkateswarlu, I. (2026). Smart Exam Malpractice Detection and Evidence Capture YOLO. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802168
Vancouver
Leelavathi C, Navya K, Yeshwanthi R, Venkateswarlu I. Smart Exam Malpractice Detection and Evidence Capture YOLO. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802168
BibTeX
@article{Leelavathi2026,
author = {Leelavathi, C. and Navya, K. and Yeshwanthi, R. and Venkateswarlu, I.},
title = {Smart Exam Malpractice Detection and Evidence Capture YOLO},
journal = {International Journal of Engineering & Extended Technologies Research},
year = {2026},
volume = {08},
number = {02},
doi = {10.15662/ijeetr.2026.0802168}
}
EUVIMED