Implementation of Driver Drowsiness Detection in Intelligent Transportation Systems Using Ai and IoT

DOI

No DOI

Published In

Volume 5, Issue 1 (2026)

Published Online

January-April 2026

Pages

400-403

Article Type

Original Article

Check for updates

No DOI available for cross-reference

Driver drowsiness is a significant contributor to road accidents worldwide, causing severe injuries, fatalities, and economic losses. To enhance road safety, this project proposes an intelligent AI-based driver drowsiness detection system integrated with Arduino for real-time vehicle intervention within Intelligent Transportation Systems (ITS). The system uses a webcam to continuously monitor the driver’s facial expressions. With Python, OpenCV, and dlib’s 68-point facial landmark detection, it extracts key facial features to compute Eye Aspect Ratio (EAR) for eye blink and closure detection and Mouth Aspect Ratio (MAR) for yawning detection. Driver drowsiness is a significant contributor to road accidents worldwide, causing severe injuries, fatalities, and economic losses. To enhance road safety, this project proposes an intelligent AI-based driver drowsiness detection system integrated with Arduino for real-time vehicle intervention within Intelligent Transportation Systems (ITS). The system uses a webcam to continuously monitor the driver’s facial expressions. With Python, OpenCV, and dlib’s 68-point facial landmark detection, it extracts key facial features to compute Eye Aspect Ratio (EAR) for eye blink and closure detection and Mouth Aspect Ratio (MAR) for yawning detection. Furthermore, the proposed system emphasizes robustness and adaptability to real-world driving environments by incorporating preprocessing techniques such as grayscale conversion, histogram equalization, and noise reduction to improve facial feature detection accuracy under varying lighting conditions, including low-light and nighttime scenarios. The CNN model is trained using a diverse dataset containing multiple facial expressions, eye states, and yawning patterns collected from different individuals to enhance generalization and reduce false detection rates. To ensure continuous performance, temporal analysis is applied to monitor consecutive frame patterns rather than relying on single-frame detection, thereby improving the reliability of drowsiness prediction. The integration of IoT-based communication allows real-time data transmission between the detection module and vehicle control unit, enabling faster response time during emergency situations. In addition, the system architecture supports modular expansion, allowing integration with cloud-based monitoring platforms for fleet management and driver behavior analytics.