PulseAugur
LIVE 04:23:15
research · [2 sources] ·
0
research

CNNs and personalized thresholds improve driver drowsiness detection accuracy

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 2 sources

Researchers have developed a new driver drowsiness detection system that uses personalized thresholds for Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) to account for individual differences. The system integrates these personalized metrics with Convolutional Neural Network (CNN) models to improve accuracy in various conditions. Evaluations showed that personalized thresholding boosted detection accuracy by 2-3%, while the CNN component achieved over 98.8% accuracy for eye state and yawning detection. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Enhances driver safety systems by improving the accuracy of fatigue detection through personalized AI models.

RANK_REASON Academic paper detailing a new method for driver drowsiness detection.

Read on arXiv cs.CV →

COVERAGE [2]

  1. arXiv cs.CV TIER_1 · G\"okdeniz Ersoy, Mehmet Alper Tatar, Eray Tonbul, Serap K{\i}rb{\i}z ·

    Improving Driver Drowsiness Detection via Personalized EAR/MAR Thresholds and CNN-Based Classification

    arXiv:2604.22479v1 Announce Type: new Abstract: Driver drowsiness is a major cause of traffic accidents worldwide, posing a serious threat to public safety. Vision-based driver monitoring systems often rely on fixed Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) thresholds; …

  2. arXiv cs.CV TIER_1 · Serap Kırbız ·

    Improving Driver Drowsiness Detection via Personalized EAR/MAR Thresholds and CNN-Based Classification

    Driver drowsiness is a major cause of traffic accidents worldwide, posing a serious threat to public safety. Vision-based driver monitoring systems often rely on fixed Eye Aspect Ratio (EAR) and Mouth Aspect Ratio (MAR) thresholds; however, such fixed values frequently fail to ge…

RELATED ENTITIES

RELATED TOPICS