Project Title: Virtual Proctor
Category: Image Processing
Project File: Download Project File
Noor Rahman
noor.rahman@vu.edu.pk
mahsud-cs619
Virtual Proctor
Project Domain
Image Processing
Introduction
The project aims to develop a simple yet effective virtual proctoring system to monitor students during online exams by detecting faces and tracking activity. With the increasing reliance on remote learning and online assessments, ensuring exam integrity has become a significant challenge. This project introduces students to the basics of face detection and real-time video monitoring. The system will detect a student’s face using the webcam, ensuring the student remains in front of the camera during the exam. If no face or multiple faces are detected, the system will alert the invigilator of potential cheating or irregular activity.
Functional Requirements:
1. Capture real-time video feed from the webcam.
2. Implement face detection using advanced methods (such as deep learning-based approaches like SSD Inception v2 or YOLOv8) to detect if a student's face is visible.
3. Detect multiple faces in the frame; raise an alert if more than one face is detected.
4. Monitor if the student leaves the frame or turns their head away for extended periods, indicating possible cheating.
5. Raise alerts (logs or messages) for unusual behavior such as no face detected, multiple faces, or the student moving out of frame.
6. Record suspicious activity with a logging system for later review.
7. Create a simple graphical interface for invigilators to monitor live video and see real-time alerts.
8. Document system design, face detection methods, experimental setup, and evaluation.
Tools & Technologies:
• Languages: Python, MATLAB
• Python:
o Libraries: OpenCV (video capture, face detection), NumPy (image handling), Matplotlib (visualization), Dlib (optional for face detection and pose estimation)
o IDE: Spyder
• MATLAB:
o Toolboxes: Image Processing Toolbox, Computer Vision Toolbox, UI
o IDE: MATLAB built-in editor
Supervisor:
Name: Noor Rahman
Email ID: noor.rahman@vu.edu.pk
Skype ID: mahsud-cs619
No reviews available for this project.