Driver Drowsiness Detection
Real-time embedded system for driver safety using Raspberry Pi with OpenCV and dlib.
- Role: Developer
- Timeline: August 2022 - November 2022
- Team: IIT Delhi Project Team (3 members)
- Technologies: OpenCV, dlib, Embedded C, Raspberry Pi
- Link: https://github.com/Laxman824/Projects-Assignments/tree/main/Realtime%20driver%20drowsiness%20detection
Problem
Drowsy driving causes 100,000+ accidents annually. Commercial solutions are expensive and not accessible. A low-cost, real-time detection system could save lives.
Solution
Built an embedded system using Raspberry Pi and camera that detects eye closure patterns (Eye Aspect Ratio) and yawning using facial landmark detection. Triggers audio alerts when drowsiness is detected.
Impact
- Real-time detection at 30 FPS
- 95% accuracy in controlled conditions
- Low-cost hardware (~$50)
- Sub-second alert latency
Key features
- Real-time face detection
- 68-point facial landmark detection
- Eye Aspect Ratio (EAR) calculation
- Yawn detection via mouth aspect ratio
- Configurable alert thresholds
- Audio alarm system
- Low-light mode with IR camera
- Data logging for analysis
Tech stack
- Hardware: Raspberry Pi 4, Pi Camera, Speaker
- Software: Python, OpenCV, dlib, NumPy
- Algorithm: Facial Landmarks, EAR Algorithm, MAR Algorithm
What K Laxman learned
- Real-time computer vision on embedded systems
- Facial landmark detection with dlib
- Optimizing OpenCV for Raspberry Pi
- Building safety-critical systems
Explore more
- Home — overview, skills and a built-in AI assistant
- Experience — roles at Think360 AI (CAMS), CAMS Mutual Funds and IIT Delhi
- Projects — GenAI, LLM, RAG and full-stack builds
- Education — IIT Delhi, M.Tech & B.Tech Computer Science
- GitHub Activity — open-source contributions
- Contact / Hire me