Personalized EAR/MAR thresholds improve drowsiness detection accuracy by 2-3% over fixed values, with CNN models reaching 99.1% accuracy on eye state and 98.8% on yawning detection.
Improving facial emotion recognition through dataset merg- ing and balanced training strategies
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Improving Driver Drowsiness Detection via Personalized EAR/MAR Thresholds and CNN-Based Classification
Personalized EAR/MAR thresholds improve drowsiness detection accuracy by 2-3% over fixed values, with CNN models reaching 99.1% accuracy on eye state and 98.8% on yawning detection.