Vision-Based PPE Detection System

Overview

The Vision-Based PPE Detection System is an AI-powered safety monitoring application designed to detect Personal Protective Equipment (PPE) compliance in workplace environments using computer vision and deep learning techniques.

The system processes uploaded video streams, analyzes frames using trained object detection models, and identifies workers who are not wearing required safety equipment such as helmets and safety goggles. Detected violations are recorded and stored for further analysis, reporting, and safety audits.

This project demonstrates the application of AI-driven safety monitoring systems for industries such as construction, manufacturing, and industrial facilities.

---

Key Features

• Automated PPE detection using trained deep learning models
• Video upload and frame processing
• Detection of missing safety equipment
• Storage of violation frames for evidence
• Alert management for safety violations
• Automated report generation
• Web-based monitoring interface

---

System Workflow

1. User uploads a surveillance video through the web interface.
2. The video processing module extracts frames from the video.
3. Each frame is analyzed using a trained object detection model.
4. PPE items such as helmets and goggles are detected.
5. If required PPE is missing, the system records a violation.
6. Violation frames are stored and reports are generated.

---

Project Structure

AI_PPE
│
├── main.py                 # Main application entry point
├── alert_manager.py       # Handles violation alerts
├── database.py            # Database connection
├── db_models.py           # Database schema models
├── report_generator.py    # Generates violation reports
├── requirements.txt       # Project dependencies
│
├── models                 # Trained detection models
│   ├── best.pt
│   └── googles.pt
│
├── services
│   └── video_processor.py # Video frame processing
│
├── static
│   ├── index.html         # Web interface
│   ├── style.css          # UI styling
│   └── script.js          # Frontend logic
│
├── uploads                # Uploaded videos
├── outputs
│   └── violations         # Detected violation images

---

Technologies Used

• Python
• YOLO (Object Detection)
• OpenCV
• HTML
• CSS
• JavaScript
• SQLite
• Bootstrap

---

Installation

1 Clone the Repository

git clone https://github.com/racervarma/PPE_Detection_System.git

2 Navigate to the Project Folder

cd PPE_Detection_System

3 Create Virtual Environment

python -m venv venv

4 Activate Environment

Windows

venv\Scripts\activate

Linux / Mac

source venv/bin/activate

5 Install Dependencies

pip install -r requirements.txt

---

Running the Application

Start the application:

python main.py

Open the web interface in your browser:

http://localhost:5000

---

Applications

• Industrial safety monitoring
• Construction site PPE compliance
• Smart factory safety systems
• Automated workplace safety audits
• Computer vision dataset generation

---

Future Enhancements

• Real-time CCTV monitoring
• Email or SMS alerts for violations
• Safety analytics dashboard
• Cloud-based deployment
• Multi-camera monitoring

---

Author

Sisir Varma

---

License

This project is developed for research and educational purposes.