Abstract
This paper proposes a method for monitoring driver safety levels using a data fusion approach based on several discrete data types: eye features, bio-signal variation, in-vehicle temperature, and vehicle speed. The driver safety monitoring system was developed in practice in the form of an application for an Android-based smartphone device, where measuring safety-related data requires no extra monetary expenditure or equipment. Moreover, the system provides high resolution and flexibility. The safety monitoring process involves the fusion of attributes gathered from different sensors, including video, electrocardiography, photoplethysmography, temperature, and a three-axis accelerometer, that are assigned as input variables to an inference analysis framework. A Fuzzy Bayesian framework is designed to indicate the driver’s capability level and is updated continuously in real-time. The sensory data are transmitted via Bluetooth communication to the smartphone device. A fake incoming call warning service alerts the driver if his or her safety level is suspiciously compromised. Realistic testing of the system demonstrates the practical benefits of multiple features and their fusion in providing a more authentic and effective driver safety monitoring.
Highlights
Drowsiness is a multidimensional feature that researchers over the past decade have found difficult to define
Dillies et al [19] used a real-time fuzzy pattern recognition process implemented in a neural network whose input is signals from the steering wheel, a sensor that measures the speed of the car, and the accelerator
Hayashi et al [24] studied ambulatory blood pressure for 24 hours in white-collar workers. They showed that subjects who worked a large amount of overtime had higher blood pressure, slept fewer hours, and were more fatigued before and after work than subjects who worked a smaller amount of overtime, whose hypertension was only mild
Summary
Drowsiness is a multidimensional feature that researchers over the past decade have found difficult to define. Proposed a real-time brain-computer interface (BCI) system to monitor human physiological and cognitive states by analyzing EEG signals. They demonstrated that the amplitude of an EEG peak value, which is estimated by a drowsiness detection system, may predict a driving error. Moreno et al [18] combined steering wheel movement data, acceleration and braking data, and speed of the vehicle for detecting that the driving pattern of a driver indicates drowsiness. Dillies et al [19] used a real-time fuzzy pattern recognition process implemented in a neural network whose input is signals from the steering wheel, a sensor that measures the speed of the car, and the accelerator. (4) a low-cost solution for capturing the driver’s image using the front-facing video sensor of a smartphone device; and (5) a reliable fake incoming call type alert system to warn and wake the driver without generating adverse effects on the driver
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
