A Real-Time Automotive Safety System Based on Advanced AI Facial Detection Algorithms
Authors:
- 2 May 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Vehicles
- Vol. 9 (6) , 5080-5100
Abstract
This research aims to design and implement a complete safety and comfort device that integrates a Driving Monitoring System (DMS) and a car Seat Electronic Control Unit (SECU). The DMS employs a combination of infrared sensors and a pre-trained Artificial Intelligence (AI) algorithm with advanced facial recognition capabilities. It aims to monitor and control the driver's distraction and drowsiness level during daytime and nighttime driving conditions by accurately detecting head pose and eye status. If the system detects the driver is drowsy or distracted from the road, a warning system will be activated by sending an electrical signal to the aforementioned car SECU. The designed SECU initiates a car seat vibration and triggers an audible signal to alert the driver of a dangerous situation by managing the seat movement process and protection features. It provides the basic drive and control of the seat motor drivers and position sensors. The most important consideration of this SECU is meeting the following requirements: real-time control of various motors, fast processing speed, low power, low heat dissipation, low quiescent current, and low Electromagnetic Interference (EMI). The two systems were integrated into a single compact device that includes a night vision camera mounted on the steering column to record data without obstructing the driver's view. The system has high precision at a maximum detection rate of 45 Frames Per Second (FPS) in real-time. Furthermore, the system is robust under actual driving conditions without being affected by the time of day, facial expressions, or eye occlusions.Keywords
Funding Information
- German Academic Exchange Service, Germany
- Wayne State University
- Office of Academic Student Affairs
- Global Engagement from the United States
This publication has 38 references indexed in Scilit:
- A review of Doherty power amplifier and load modulated balanced amplifier for 5G technologyInternational Journal of Circuit Theory and Applications, 2023
- Eye fatigue estimation using blink detection based on Eye Aspect Ratio Mapping(EARM)Cognitive Robotics, 2022
- Fuzzy-based Driver Monitoring System (FDMS): Implementation of two intelligent FDMSs and a testbed for safe driving in VANETsFuture Generation Computer Systems, 2020
- Drowsiness and distraction while driving: A study based on smartphone app dataJournal of Safety Research, 2020
- Driver Behavior Recognition in Driver Intention Inference Systems2020
- Social Norms and Risk Perception: Predictors of Distracted Driving Behavior Among Novice Adolescent DriversJournal of Adolescent Health, 2014
- Head Pose Estimation by Perspective-n-Point Solution Based on 2D Markerless Face Tracking2014
- Impact of distracted driving on safety and traffic flowAccident Analysis & Prevention, 2013
- Model-Based Pose Estimation2011
- An overview of the polymer-to-metal direct-adhesion hybrid technologies for load-bearing automotive componentsJournal of Materials Processing Technology, 2008