About My Project

Project Title

Non-Contact AI-Drowsiness Detection System for Safe Driving

Problem

Driver drowsiness is a critical issue contributing to accidents, severe crashes, and fatalities on highways. Existing solutions often require physical contact or are not efficient enough for real-time monitoring. There is a need for a non-contact, reliable system that can accurately detect driver drowsiness and provide timely alerts.

Approach

This project will develop a non-contact AI-based drowsiness detection system using the following steps:

  • Image Collection: Use cameras to capture video feeds of the driver’s face and eye movements.
  • Image and Video Analysis: Employ computer vision techniques to analyze the collected images and videos for signs of drowsiness.
  • Deep Learning Modeling: Develop predictive models using machine learning algorithms, including transfer learning, to detect drowsiness indicators.
  • Real-Time Monitoring System: Integrate the predictive model into a real-time system that processes images and videos to detect early signs of drowsiness.
  • Alert System: Create an alert system that notifies the driver through audio cues when drowsiness is detected.

Expected Outcome

By the end of the program, we will develop a robust non-contact AI-based drowsiness detection system. We will utilize ensemble modeling to combine multiple neural network models, harnessing the strengths of each to create a more powerful cumulative model. Our system will monitor eye and facial expressions using images and live-feed videos to detect signs of drowsiness, sending an alert signal to the driver if drowsiness is detected. With this project, we aim to make driving safer by reducing the risk of accidents caused by drowsy driving. We plan to build on existing research and will quantify our system’s performance using metrics such as F1 score, recall, and precision. This project will culminate in a professional research paper to be published in a reputable journal, with us undergraduates as co-authors.

Final Report: View PDF

Graduate Student Mentor

Pelumi Olaitan Abiodun

Faculty Mentor

Dr. Oludare Owolabi