Driver Posture Recognition: A Review

Abdul S. Mahomed

doi:10.1109/ACCESS.2024.3496578

2024-01-01

Abdul S. Mahomed

Akshay Kumar Saha *

DOI：10.1109/ACCESS.2024.3496578

原文链接

原文求助

摘要

En 中文

Driver Posture Recognition Systems (DPRS) are indispensable for advancement in vehicle safety by tracking driver posture using sensors and cameras. These systems monitor the driver's position, movements, and posture to promote safe driving behaviour and take action if needed using a variety of approaches to driver posture analysis. Sensor based approaches involved the utilisation of sensors such pressure sensors, accelerometers, and gyroscopes, to collect information on the driver's posture when seated in vehicle. Vision based approaches consisted mainly of cameras to monitor and assess the driver's position using advanced computer vision methods associated with facial landmark detection and body posture analysis. Hybrid approaches is the merging of both sensor-based and vision-based approaches to provide improved precision through the integration of various data sources using fusion tactics, feature extraction methods, and machine learning algorithms. The implementation of DPRS in vehicles is centred on monitoring driver posture in real-time without intrusion, providing prompt alerts to reduce risks related to incorrect seating or positioning. This article provides a general review of most aspects related to DPRS design, emphasizing the significance of accurate posture recognition that give priority to driver health and safety.

Keyword:

Vehicles

Sleep

Fatigue

Sensors

Monitoring

Feature extraction

Accuracy

Visualization

Sensor systems

Face recognition

Autonomous vehicles