Abstract:Drowsiness detection is essential for improving safety in areas such as transportation and workplace health. This study presents a real-time system designed to detect drowsiness using the Eye Aspect Ratio (EAR) and facial landmark detection techniques. The system leverages Dlibs pre-trained shape predictor model to accurately detect and monitor 68 facial landmarks, which are used to compute the EAR. By establishing a threshold for the EAR, the system identifies when eyes are closed, indicating potential drowsiness. The process involves capturing a live video stream, detecting faces in each frame, extracting eye landmarks, and calculating the EAR to assess alertness. Our experiments show that the system reliably detects drowsiness with high accuracy while maintaining low computational demands. This study offers a strong solution for real-time drowsiness detection, with promising applications in driver monitoring and workplace safety. Future research will investigate incorporating additional physiological and contextual data to further enhance detection accuracy and reliability.

What problem does this paper attempt to address?

The main objective of this paper is to develop a real-time driver fatigue detection system by calculating the Eye Aspect Ratio (EAR) and using facial landmark detection technology. Specifically, the research goals include: 1. **Create a real-time system**: Develop a system capable of detecting fatigue in real-time, based on the detection of the Eye Aspect Ratio and facial landmarks. 2. **Evaluate system accuracy**: Assess the ability of the developed system to recognize signs of fatigue in real-time video streams. 3. **Determine optimal thresholds**: Identify the best EAR threshold and frame inspection parameters for reliable fatigue detection. 4. **Performance comparison**: Compare the proposed approach with existing methods, particularly those aimed at driver fatigue detection. 5. **Limitations and future directions**: Identify potential limitations and suggest further research directions to enhance system performance. The study utilizes advanced facial recognition technologies, such as pre-trained shape predictor models, to accurately detect and monitor 68 facial landmarks. These landmarks are used to calculate the EAR, and by setting an EAR threshold, the system can identify eye closure, indicating potential fatigue. The system is capable of reliably detecting fatigue while maintaining low computational demands, making it suitable for practical applications such as driver monitoring and workplace safety. Experimental results show that the system exhibits high accuracy, precision, and recall under various conditions. It performs best in well-lit environments. However, the system's performance decreases under low-light or nighttime conditions, indicating that future research needs to address how to improve the system to adapt to different lighting environments. Additionally, the study suggests several future research directions, such as improving techniques for handling occlusions and rapid head movements, personalized calibration, and the integration of multimodal data to further enhance the system's robustness and accuracy.

Real-Time Drowsiness Detection Using Eye Aspect Ratio and Facial Landmark Detection

Real Time Implementation of Driver Drowsiness Detection for an Intelligent Transportation System

VigilEye -- Artificial Intelligence-based Real-time Driver Drowsiness Detection

Eye behaviour based drowsiness Detection System

In-the-wild Drowsiness Detection from Facial Expressions

Real-Time System for Driver Fatigue Detection by RGB-D Camera

Joint Shape And Local Appearance Features For Real-Time Driver Drowsiness Detection

Driver Drowsiness Recognition Based on Computer Vision Technology

Driver Drowsiness Detection using Machine Learning and Deep Learning

A sophisticated Drowsiness Detection System via Deep Transfer Learning for real time scenarios

A Real-Time Driving Drowsiness Detection Algorithm With Individual Differences Consideration

Driver Drowsiness Detection using Artificial Intelligence

Towards a new system for drowsiness detection based on eye blinking and head posture estimation

Survey Paper for Real Time Car Driver Drowsiness Detection using Machine Learning Approach

Driver Drowsiness Detection System: An Approach By Machine Learning Application

Towards Real-Time Drowsiness Detection for Elderly Care

Drowsy Driving Detection System Using Face Detection

Real-Time Fatigue Detection Algorithms Using Machine Learning for Yawning and Eye State

SleepyWheels: An ensemble model for drowsiness detection leading to accident prevention

DrowsyDetectNet: Driver Drowsiness Detection Using Lightweight CNN With Limited Training Data

Design of Real-Time Driver Drowsiness Detection Based Electromechanical Braking System for Heavy-Duty Vehicles