Advanced Driver & Road Safety System
EyeOnDriver (ADAR 1.0) is an intelligent, real-time driver monitoring system designed to enhance road safety using AI-powered computer vision and IoT hardware integration. Built with Python (OpenCV & MediaPipe) and ESP32 microcontroller, this system proactively detects unsafe driver behaviors and environmental conditions, preventing accidents through early alerts and control actions.
It is designed to be low-cost, open-source, and scalable for deployment in personal vehicles, commercial fleets, and public transport—especially across developing regions.
| Feature | Description |
|---|---|
| 👁️ Drowsiness Detection | Detects closed eyes via Eye Aspect Ratio (EAR) using face landmarks. |
| 📱 Phone Usage Detection | Identifies hand-to-ear gesture or downward gaze using MediaPipe landmarks. |
| 🗣️ Talking Detection | Tracks continuous mouth motion to flag phone conversation or distraction. |
| 🧑 Facial Recognition | Only authorized drivers can start/operate the vehicle. |
| 🧾 Aadhaar Age Verification | Prototype simulation checks for 18+ eligibility to drive. |
| 🍷 Alcohol Detection (MQ-3) | ESP32 + gas sensor prevents ignition if alcohol is detected. |
| 🔊 Voice Alerts (TTS) | Speaks real-time warnings via text-to-speech. |
| 🧠 ESP32 Integration | Communicates with Python via serial to display messages, trigger buzzers. |
| 🆘 SOS Feature (Planned) | Auto-alert + location tracking for emergencies (future scope). |
- Python: OpenCV, MediaPipe, pyttsx3, serial
- Hardware: ESP32, MQ-3 sensor, 16x2 LCD (I2C), buzzer, LEDs
- Tools: Arduino IDE, VS Code, Fritzing (for circuits)
- Future Additions: Firebase/ThingsBoard dashboard, GPS + GSM integration
[ Laptop Camera ]
↓
[ Python + OpenCV ]
↓
[ Detection Logic ]
↓
[ Serial Communication ]
↓
[ ESP32 Board ]
↓ ↓ ↓
[Buzzer] [LCD Display] [Relay Lock]
EyeOnDriver/
├── README.md
├── requirements.txt
├── /python_code
│ └── driver_monitor.py
├── /esp32_code
│ └── esp32_driver_alert.ino
├── /images
│ └── demo_setup.jpg
├── /circuit_diagrams
│ └── wiring_schematic.pdf
├── /docs
│ └── project_description.pdf
└── /videos
└── demo_clip.mp4