SafeDrive – Real-Time Drowsiness Detection System

A real-time driver drowsiness detection system using Mediapipe Face Mesh, OpenCV, and EAR/MAR-based alerting.

---

🚀 Features

Feature	Description
👁️ Real-time Eye Tracking	Uses Mediapipe Face Mesh to track eye landmarks frame-by-frame
😴 Drowsiness Detection (EAR)	Detects prolonged eye closure using Eye Aspect Ratio
🗣️ Yawn Detection (MAR)	Identifies yawning patterns using Mouth Aspect Ratio
🔊 Instant Alerts	Triggers audio alarms on detecting drowsiness
🎥 Streamlit Web App	Live, browser-based detection using webcam

---

📸 Demo

(Add gif or images here)

demo.gif

---

📁 Project Structure

driver-drowsiness-detection-system
│── drowsy_detection.py
│── audio_handling.py
│── streamlit_app.py
│── requirements.txt
│── README.md
│
├── audio/
│   └── wake_up.wav
│
├── test-open-eyes.jpg
└── test-close-eyes.jpg

---

🛠️ Tech Stack

• Python
• OpenCV
• Mediapipe Face Mesh
• Streamlit
• streamlit-webrtc
• pydub / av

---

⚙️ Installation

1️⃣ Clone the repository

git clone https://github.com/vatsalm1611/driver-drowsiness-detection-system
cd driver-drowsiness-detection-system

2️⃣ Install dependencies

pip install -r requirements.txt

---

▶️ Run the System

Option A — Run Streamlit Web App

streamlit run streamlit_app.py

Option B — Run the basic webcam EAR monitor

python drowsy_detection.py

---

🔍 How It Works

Eye Aspect Ratio (EAR)

• Detects eye openness using 6 facial landmark points.
• If EAR < threshold for continuous duration → Drowsiness detected.

Mouth Aspect Ratio (MAR)

• Detects wide mouth opening (yawn).
• MAR > threshold → Yawn event.

---

🎯 Key Logic (Simplified)

Drowsiness = EAR below threshold for X seconds
Yawn = MAR above threshold

Both metrics are computed per frame → real-time detection.

---

🤝 Contributing

Contributions are welcome!

1. Fork the repo
2. Create a new branch
3. Commit your changes
4. Open a Pull Request

---

📝 License

This project is licensed under the MIT License.

---

🙏 Acknowledgements

• Mediapipe team
• OpenCV community
• Streamlit contributors

---

If you like this project, ⭐ the repository!