This project was developed as part of a college academic project, demonstrating the practical application of computer vision and machine learning concepts in real-world safety systems.
-
Enhanced Detection
- Head pose estimation for distraction detection
- Machine learning for personalized drowsiness patterns
- Integration with cabin temperature and humidity sensors
-
Vehicle Integration
- Direct CAN bus integration for vehicle control
- Integration with existing autonomous systems
- Fleet management system connectivity
-
Safety Enhancements
- GPS integration for safe parking locations
- Emergency services notification
- Cloud-based driver monitoring analytics
-
Hardware Improvements
- Infrared camera support for night operation
- Multiple camera angles
- Integration with steering wheel sensors
- Vital sign monitoring through steering wheel sensors
-
Mobile Integration
- Real-time fleet manager notifications
- Mobile app for system monitoring
- Driver performance analytics
Special thanks to the MediaPipe team for their excellent face detection and landmark tracking solutions that made this project possible. This project builds upon their work in computer vision and machine learning.
This system is a proof of concept and should not be used as the sole means of ensuring driver safety. Always follow local regulations and safety guidelines regarding commercial vehicle operation.
This project is licensed under the MIT License
- Bezalelohim
- SharanSundar
- v1.0.0 (December 2022) - Initial release
- v1.0.1 (December 2022) - Performance optimizations for Raspberry Pi