Table 1.

Approaches for detecting driver’s drowsiness.

Methods	Measurement Information	Measurement Target	Measurement Method	Measurement Index	Advantages	Disadvantages
Contact method	Biometric information	Heartbeat, pulse wave, aspiration, brain wave, myoelectric, eye movement, etc.	Heart rate monitor, pulse wavemeter, electroencephalograph, electromyograph, nystagmus, etc.	Heart rate, chaos analysis, alpha wave, theta wave, muscle action potential, vestibular oculomotor reflex, etc.	High drowsiness detection performance can be obtained [10].	Driver behavior adversely affects the reliability of the designed system [10].
Non-contact method	Vehicle behavior	Steering pattern, distance between lane and vehicle, speed, distance between vehicles, etc.	Steering angle sensor, white line recognition camera, laser radar, etc.	Steering frequency, meandering rate, steering volume, monotonous steering	Estimates can be obtained in a way that is unobtrusive to the driver [11].	The accuracy of detection and estimation depends on road conditions and the environment. It is useful only when the driver is holding the steering wheel.
	Driver’s graphic information	Open rate of eyes, blink, pupil, voice, and expression	Camera, microphone	Opening and closing rates of eyes, number of blinks, time of closing eyes, pupil fluctuation, chaos analysis of speech sound, drowsy categorization by expression, etc.	Intuitional index and easy-to-understand, high accuracy.	Different lighting conditions may disrupt the detection performance [10].