DriverSense AI - Drowsiness Detection & Alert System

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for DriverSense AI - Drowsiness Detection & Alert System

Inspiration

Fatigue and microsleep while driving are among the leading causes of highway accidents. We wanted to build a lightweight, accessible AI-driven safety assistant that continuously monitors driver alertness and triggers instant auditory warnings to prevent crashes.

What it does

DriverSense AI tracks facial landmarks in real time through a live camera feed. It monitors the Eye Aspect Ratio (EAR) and head posture. If the driver closes their eyes for consecutive frames or nods off, the system immediately plays a loud alert sound to wake the driver.

How we built it

• Computer Vision & AI: Built using Python, OpenCV, and facial landmark models to calculate Eye Aspect Ratio (EAR).
• Alert Mechanism: Programmed an audio playback module to sound an alarm upon detecting fatigue thresholds.
• Frontend/Interface: Built an intuitive dashboard using modern web tools and JavaScript for real-time video feedback.

Challenges we faced

Optimizing the detection pipeline to run with low latency and avoiding false alarms caused by normal blinking was challenging. We fine-tuned the threshold frames to ensure alerts trigger only on prolonged eye closure.

What's next

Adding yawning detection, heart-rate estimation, and deploying it on embedded edge hardware like Raspberry Pi for direct car dashboard integration.

Built With

• computer-vision
• html5
• javascript
• opencv
• python