https://github.com/darshan1654/ai-ppe-detection
An AI-powered CCTV surveillance system for real-time detection of PPE compliance, including helmet and mask violations, using YOLO and computer vision.
https://github.com/darshan1654/ai-ppe-detection
cctv-surveillance industrial-safety opencv ppe-detection python streamlit yolo
Last synced: about 2 months ago
JSON representation
An AI-powered CCTV surveillance system for real-time detection of PPE compliance, including helmet and mask violations, using YOLO and computer vision.
- Host: GitHub
- URL: https://github.com/darshan1654/ai-ppe-detection
- Owner: darshan1654
- License: mit
- Created: 2025-07-14T15:08:17.000Z (12 months ago)
- Default Branch: main
- Last Pushed: 2025-07-14T19:34:01.000Z (12 months ago)
- Last Synced: 2025-07-14T22:18:37.965Z (12 months ago)
- Topics: cctv-surveillance, industrial-safety, opencv, ppe-detection, python, streamlit, yolo
- Language: Python
- Homepage: https://ai-ppe-detection-dp.streamlit.app/
- Size: 17 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🛡️📹 AI-Powered Industrial Safety Surveillance System
---
**Category:** Industry Defined Problem | **Organization:** Jyoti CNC Automation, Rajkot
 
---
## 📄 Project Overview
This project integrates **AI-powered Personal Protective Equipment (PPE) detection** and **CCTV-based Anomaly Surveillance** into a single real-time system. It is designed to improve **Workplace Safety**, **Regulatory Compliance**, and **Operational Efficiency** in industrial environments.
Using advanced **Deep Learning models** and **Computer Vision**, the system ensures that workers adhere to safety protocols and that unusual or unsafe activities are promptly flagged.
---
## ✅ Key Features
### 👷 PPE Detection Module
- Real-time detection of:
- 🪖 Helmets
- 😷 Face Masks
- 👷 Safety Vests
- 🧤 Gloves
- Supports live feed from **Webcam/IP Camera**
- Upload and analyze **Video/Image files**
- Detection logs with timestamps and confidence scores
### 📹 CCTV Anomaly Detection Module
- Real-time detection of:
- 🚫 Entry into restricted zones
- ⚠️ Safety violations (e.g., no helmet, improper behavior)
- 🚷 Suspicious or unsafe movements
- Continuous monitoring via CCTV/IP camera
- Alert generation on detection
### 📊 Dashboard (Built with Streamlit)
- Live status feed with detection results
- Real-time preview of camera feed
- Violation and compliance logs
- Summary statistics and compliance reports
---
## 📦 Tech Stack
| Component | Technology |
|----------------|--------------------|
| 💡 AI Model | YOLOv8 (Ultralytics) |
| 🧠 Backend | Python, OpenCV |
| 🌐 Frontend/UI | Streamlit |
| 🎥 Video Input | Webcam/IP Camera |
| 📊 Data Logging | Pandas, CSV Logs |
---
## 🏭 Industrial Benefits
- ✅ Automated compliance with PPE policies
- 🔍 Real-time safety monitoring
- 📉 Reduced accident risk and manual supervision
- 📊 Actionable insights from safety data
---
## 📸 Sample Outputs
- 📷 Detected image with PPE boxes
- 🧾 Logs with timestamp and violation type
- 📈 Streamlit dashboard with real-time updates
---
## 🔧 Installation & Setup
```bash
# 1. Clone the repository
git clone https://github.com/darshan1654/AI-PPE-Detection.git
# 2. Create and activate a virtual environment
python -m venv venv
source venv/bin/activate # or venv\Scripts\activate on Windows
# 3. Install required dependencies
pip install -r requirements.txt
# 4. Run the application
streamlit run app.py
```
---
## 📁 Project Structure
```
├── app.py # Streamlit app
├── best.pt # YOLOv8 model files
├── yolov8n.pt # YOLOv8 model files
├── violation_logs.csv # Detection logs CSV file
├── requirements.txt # Dependencies
├── packages.txt # packages
└── runtime.txt # For deployment (e.g., Heroku)
```
---
## 👨💻 Team Details
**Group ID:** G00171
| Name | Email |
| ---------------- | --------------------------------------------------------------------------- |
| Kushal A. Parekh | [22ce113@svitvasad.ac.in](mailto:22ce113@svitvasad.ac.in) |
| Darshan Pardeshi | [darshanpardeshi1654@gmail.com](mailto:darshanpardeshi1654@gmail.com) |
| Param V. Jani | [janiparam61@gmail.com](mailto:janiparam61@gmail.com) |
| Darshan Panchal | [mpdarshanpanchal001031@gmail.com](mailto:mpdarshanpanchal001031@gmail.com) |
| Jaymin Raval | [ravaljaymin2908@gmail.com](mailto:ravaljaymin2908@gmail.com) |
---
## 🔮 Future Scope
* 🔔 Voice/Email/SMS alert system
* 🔗 Integration with ERP systems
* 📊 Admin dashboard with analytics
* 📤 Auto-upload violation clips
* 📡 Multi-location camera support
---
## 🏆 Achievements
* ✅ Used in **Real Industrial setup** at Jyoti CNC
* 🎓 Presented at **College-level expo**
* 📡 Successfully tested with **Live Camera feeds**
---
## 📜 License
For **Academic and Research use** only.
---
## 🙏 Acknowledgements
* [Ultralytics YOLOv8](https://github.com/ultralytics/ultralytics)
* [OpenCV](https://opencv.org)
* [Streamlit](https://streamlit.io)
* Jyoti CNC Automation Pvt. Ltd.
* SVIT Vasad
---
## 📬 Contact
📧 [janiparam61@gmail.com](mailto:janiparam61@gmail.com)
📧 [darshanpardeshi1654@gmail.com](mailto:darshanpardeshi1654@gmail.com)
📧 [22ce113@svitvasad.ac.in](mailto:22ce113@svitvasad.ac.in)
📧 [mpdarshanpanchal001031@gmail.com](mailto:mpdarshanpanchal001031@gmail.com)
📧 [ravaljaymin2908@gmail.com](mailto:ravaljaymin2908@gmail.com)
---