StayAwake

Detects if long-haul drivers are falling asleep, plays an alarm if they are drowsy, and gathers data for analytics and data-driven logistic planning.

Comment 2

Inspiration

According to the NSC, roughly 18% of truck drivers report falling asleep behind the wheel, and drowsy driving could be involved in upwards of 40% of truck crashes. Drowsiness while driving is hazardous for carriers and surrounding drivers. Keeping carriers awake and alert could save lives. We hope to develop a product to ensure the safety of carriers behind the wheel and provide analytics to avoid driving fatigue by smarter route and schedule planning .

What it does

Real-time face mesh detection and landmark extraction using OpenCV2 and mediapipe->Eye aspect ratio (EAR) calculated per frame indicates drowsiness, triggering alarm when needed->Storage of geographical information (lat. & long.) in SQL server where carriers frequently fall asleep->Arrange stops before high-risk segments or adjust schedule accordingly

How we built it

App development in Flask, HTML, and CSS. Eye recognition and drowsiness detection using OpenCV2 and Mediapipe. Data storage in Microsoft Azure SQL Server database.

Challenges we ran into

Frontend and UI design, machine learning model deployment

What's next for Stay Awake

Stay tuned!

Built With

Share this project:

Updates