Skip to content

Guard AI is a real-time AI-based remote proctoring system that ensures secure, fair online assessments. It tracks website activity, detects speaking, and monitors user focus during exams, generating detailed session reports. Built with a modern frontend and a Node.js backend, Guard AI aims to redefine trust in online evaluations.

License

Notifications You must be signed in to change notification settings

Nikhil210206/Mozohack_Guard-AI

Repository files navigation

Guard AI

image

Welcome to the Guard AI project !! 🚨

Guard AI is a lightweight, AI-powered exam monitoring tool that provides real-time proctoring by detecting suspicious activities during online exams such as unauthorized speaking, gaze deviation, and usage of restricted websites.

🔍 Features :-

Lip Movement Detection + Audio Analysis Detects if the user is speaking during the exam. Differentiates between yawning and speaking.

Screenshot 2025-04-22 080606

👀 Gaze Tracking :-

Tracks user's eye and head movement. Detects if the user looks away from the screen.

Screenshot 2025-04-22 080633

🖥️ Website Monitoring :-

Detects if the user opens any non-permitted websites. A warning signal is provided if done so.

image

📂 Comprehensive Report Generation :-

Summarizes user behavior during the exam. Includes start and end times of suspicious activities. Generates a clean PDF report after each test session.

📚 Project Structure :-

guard_ai_project/ │ ├── features/ │ ├── init.py │ ├── lip_detection.py │ ├── gaze_tracking.py │ └── website_monitor.py │ ├── logs/ │ ├── lip_audio_logs.txt │ ├── gaze_tracking_logs.txt │ └── website_usage_logs.txt │ ├── reports/ │ └── final_report.pdf (auto-generated) │ ├── main.py ├── generate_report.py ├── README.md ├── requirements.txt

📂 Comprehensive Report Generation :-

🔧 **Installation** :- 1. Clone the repository: git clone https://github.com/your-username/guard-ai-project.git 2. Install the required libraries: pip install -r requirements.txt

📅 How to Use :-

1. Run the application: python main.py
  1. After finishing, you can generate an intermediate report.

  2. At the end, generate a final comprehensive PDF report.

📊 Reports :-

Reports will be generated automatically in the reports/ folder.

The report includes: Activities detected. Time intervals. Observations.

🎉 Credits :

Developed by Nikhil Balamurugan and Vishaal Pillay👨‍💻

🌍 License :

This project is licensed under the MIT License.

About

Guard AI is a real-time AI-based remote proctoring system that ensures secure, fair online assessments. It tracks website activity, detects speaking, and monitors user focus during exams, generating detailed session reports. Built with a modern frontend and a Node.js backend, Guard AI aims to redefine trust in online evaluations.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •