Published Paper Details:

Accidents caused by fatigue are a serious problem, particularly in sectors like manufacturing and transportation. This work suggests a deep learning-based method for face recognition-based driver drowsiness detection and accident prevention. Driver fatigue is a major cause of road accidents, making real-time detection crucial for ensuring safety. Traditional methods like K-Nearest Neighbors (KNN) classify fatigue based on facial features but struggle with real-time tracking and temporal dependencies. To overcome these limitations, this project introduces a CNN + LSTM-based fatigue detection system that not only extracts facial features automatically but also analyzes fatigue patterns over time for improved accuracy. The system employs a Convolutional Neural Network (CNN) for feature extraction from facial images and a Long Short-Term Memory (LSTM) network to analyze temporal patterns associated with fatigue indicators such as eye closure. The proposed model processes real-time video frames, detects fatigue symptoms, and triggers alerts to prevent drowsy driving. The system is designed to be efficient and scalable, making it suitable for real-world deployment in vehicles and industrial safety applications. Experimental results demonstrate high accuracy in fatigue detection, showcasing the effectiveness of combining CNN and LSTM for robust facial-based fatigue analysis.