INTELLIGENT DRIVER DROWSINESS DETECTION THROUGH REAL-TIME EYE MONITORING
Keywords:
Driver Drowsiness Detection, Road Safety, Driver Monitoring System, Smart Vehicle Systems, CNN, RNN, LSTM, Real-Time Monitoring, Deep Learning, Eye State Identification, Behavioral AnalysisAbstract
This paper presents a novel architecture for driver drowsiness detection that leverages real-time eye state recognition from live video feeds. The system achieves an accuracy of 97% and is capable of issuing timely alerts to help prevent accidents caused by driver fatigue—one of the major contributors to traffic-related fatalities worldwide. The proposed framework integrates Convolutional Neural Networks (CNNs) with Recurrent Neural Networks (RNNs) to enhance feature extraction and temporal sequence analysis. A comprehensive dataset of 4,760 images, evenly divided between open-eye (2,380) and closed-eye (2,380) states, captured under diverse driving conditions, was used to train and validate the model. The results demonstrate the effectiveness of the hybrid deep learning approach in improving detection accuracy, offering a reliable and scalable solution for enhancing road safety.
