Article · Engineered Science · January 1, 2026

A Hybrid Computer Vision Framework for Real-Time Driver Fatigue and Distraction Detection Using Deep Learning

Indira Uvaliyeva, Bauyrzhan Smakanov

Abstract

Driver fatigue and distraction are critical behavioral states that require reliable, non-contact, and computationally efficient methods for real-time monitoring.This study presents a hybrid computer vision framework that combines interpretable geometric features with deep-learning-based object detection.The framework integrates Eye Aspect Ratio (EAR), Percentage of Eye Closure (PERCLOS), Mouth Aspect Ratio (MAR), head pose estimation, gaze direction analysis, and convolutional neural network (CNN)-based mobile phone detection within a unified multimodal architecture.A

Sleep and Work-Related FatigueHuman-Automation Interaction and SafetyFace recognition and analysisdriver fatigue detectiondriver distraction detectionreal-time detection

A Hybrid Computer Vision Framework for Real-Time Driver Fatigue and Distraction Detection Using Deep Learning

Abstract

More by Bauyrzhan Smakanov