The Driver Fatigue Detection

Published Date: 02 Nov 2017

The main idea behind this project is to develop a nonintrusive system which can detect fatigue of the driver and issue a timely warning. Since a large number of road accidents occur due to the driver drowsiness. Hence this system will be helpful in preventing many accidents, and consequently save money and reduce personal suffering. This system will monitor the driver’s eyes using camera and by developing an algorithm we can detect symptoms of driver fatigue early enough to avoid accident. So this project will be helpful in detecting driver fatigue in advance and will gave warning output in form of sound and seat belt vibration whose frequency will vary between 100 to 300 Hz. Moreover the warning will be deactivated manually rather than automatically. So for this purpose a deactivation switch will be used to deactivate warning. Moreover if driver felt drowsy there is possibility of sudden acceleration or de acceleration hence we can judge this by Plotting a graph in time domain and when all the three input variables shows a possibility of fatigue at one moment then a Warning signal is given in form of text or red color circle. This will directly give an indication of drowsiness/fatigue which can be further used as record of driver performance.

Keywords: Fatigue Detection, Fatigue Warning, Eye tracking system, Real time fatigue detection, Driver monitoring system.

Introduction

The Real Time dangerous behaviors which are related to fatigue whether in form of eye closing, head nodding or the brain activity. Hence we can either measure change in physiological signals, such as brain waves, heart rate and eye Blinking or by measuring physical changes such as sagging posture, leaning of driver’s head and open/closed state of eyes.

The previous technique, while more accurate, is not realistic since highly sensitive electrodes would have to be attached directly on the driver’s body and hence which can be annoying and distracting to the driver. In addition long time driving would result in perspiration on the sensors, diminishing their ability to monitor accurately. The second technique is to measure physical changes (i.e. open/closed eyes to detect fatigue) is well suited for real world driving conditions since it is non-intrusive by using a video camera to detect changes. In addition micro sleeps that are short period of sleeps lasting 2 to 3 seconds are good indicator of fatigue state. Thus by continuously monitoring the eyes of the driver one can detect the sleepy state of driver and timely warning is issued.

The driver fatigue detection based on tracking the mouth and to study on monitoring and recognizing yawning. The authors proposed a method to locate and track driver’s mouth using cascade of classifiers proposed by Viola-Jones for faces. SVM is used to train the mouth and yawning images. During the fatigue detection mouth is detected from face images using cascade of classifiers. Then, SVM is used to classify the mouth and to detect yawning then alert Fatigue was described [1].

To provide reliable indications of driver drowsiness based on the characteristics of driver–vehicle interaction. A test bed was built under a simulated driving environment, and a total of 12 subjects participated in two experiment sessions requiring different levels of sleep (before the experiment. The performance of the subjects was analyzed in a series of stimulus-response and routine driving tasks, which revealed the performance differences of drivers under different sleep deprivation levels. In addition, they presented both qualitative and quantitative guidelines for designing drowsy-driver detection systems in a probabilistic frame work based on the paradigm of Bayesian networks [2].

The approaches are divided by the following five different types of measures: 1) subjective report measures; 2) driver biological measures; 3) driver physical measures; 4) driving performance measures; and 5) hybrid measures. They also discuss some nonlinear modeling techniques commonly used in the literature was summarized [3].

The reviews and compares current status of research in modeling fatigue where fatigue is modeled using probabilistic models, machine learning models, finite state machine etc. The paper also presents possible future research directions in the same field like identifying non-fatigue non-vigilance mental states, extending non-vigilance monitoring for mass audience ET care discussed [4].

An automatic drowsy driver monitoring and accident prevention system that is based on monitoring the changes in the eye blinks duration. Our proposed method detects the drowsiness in eyes using the proposed mean sift algorithm. Our new method detects eye blinks via a standard webcam in real-time YUY2_640x480 resolution. Experimental results in the eye-blink database showed that the proposed system detects eye blinks with 99.4% accuracy with a 1% false positive rate described [5].

An artificial neural network to detect the driver drowsiness was introduced [6] they an "eye mouse" to provide computer access for people with severe disabilities. The proposed eye mouse allows people with severe disabilities to use their eye movements to manipulate computers. It requires only one low-cost Web camera and a personal computer. A five stage algorithm is developed to estimate the directions of eye movements and then use the direction information to manipulate the computer. Several experiments were conducted to test the performance of the eye tracking system was implemented [7].

An IR camera is placed in front of the driver, in the dashboard, in order to detect his face and obtain drowsiness clues from their eyes closure. It works in a robust and automatic way, without prior calibration. The presented system is composed of 3 stages. The first one is pre-processing, which includes face and eye detection and normalization. The second stage performs pupil position detection and characterization, combining it with an adaptive lighting filtering to make the system capable of dealing with outdoor illumination conditions. The final stage computes PERCLOS from eyes closure information was described [8].

A review of Intelligent Transport System for motorcycles safety and related issues with some existing or emerging ITS technologies to enhanced vehicles safety. Intelligent Transport Systems (ITS) have significant potential to enhance traffic safety. Numerous ITS technologies have been developed to improve the safety and efficiency of cars, commercial vehicles, public transport and infrastructure. [9].

A real-time non-intrusive prototype driver fatigue monitor. It uses remotely located CCD cameras equipped with active IR illuminators to acquire video images of the driver. Various visual cues typically characterizing the alertness of the driver are extracted in real time and systematically combined to infer the fatigue level of the driver. The visual cues employed characterize eyelid movement, gaze movement, head movement, and facial expression was described [10].

Problem Definition

The technique used previous is more accurate, but it is not realistic since highly sensitive electrodes would have to be attached directly on the driver’s body and hence which can be annoying and distracting to the driver. In addition long time driving would result in perspiration on the sensors, diminishing their ability to monitor accurately. The second technique is to measure physical changes (i.e. open/closed eyes to detect fatigue) is well suited for real world driving conditions since it is non-intrusive by using a video camera to detect changes. In addition micro sleeps that are short period of sleeps lasting 2 to 3 seconds are good indicator of fatigue state. Thus by continuously monitoring the eyes of the driver one can detect the sleepy state of driver and timely warning is issued.

Flowchart

The Algorithm used here is very much fast as compared to PERCLOS[5] earlier used by other hence the Processing time of this system is less than half second hence it is quite fast and issues timely warning to the driver.

This system will detect a driver fatigue by processing of eye-region. As shown in flow chart in Figure 1. After image acquisition, face detection is the first stage of processing. Then symptoms of hypo-vigilance are extracted from the eyes. If eyes are blinking normally no warning is issued but when the eyes are closed for more than half second this system issues warning to the driver in form of alarm and vibration.

Driver’s image

Face detection

Eye detection

Recognition of eyes whether

Open or closed

Calculation of criteria for judging

Drowsiness/ fatigue

Is driver

Drowsy

Sound\seat belt vibration warning

No

Yes

Figure 1: Eye tracking system

Block Diagram:

As shown in Figure 2 this system consists of three input variables (i.e. Steering wheel gripping pressure variability detection unit, Speed variation detection unit and Eye Tracking Unit) and one Image processing unit (i.e. computer) and two output units (i.e. Seat belt Vibration warning and Audio alarm) and a microcontroller.

PERCLOS (PERcent eyelid CLOSure) is a measure of driver alertness, which was identified as the most reliable and valid in a study by the US Federal Highway Administration; various authors refer PERCLOS as a standard for drowsiness detection. The measure is the percentage of eyelid closure over the pupil over time and reflects slow eyelid closures rather than blinks. The PERCLOS drowsiness metric was established in a 1994 driving simulators study as the proportion of time (%) in a minute when the eyelids are at least 80 percent closed. Based on the research by Wierwill, the US Federal Highway Administration (FHWA) and US National Highway Traffic Safety Administration (NHTSA) consider PERCLOS as among the most promising known real time measures of alertness for in-vehicle drowsiness detection systems.

P70: the proportion of time when the eyes were closed at least 70 percent.

P80: the proportion of time when the eyes were closed at least 80 percent (the P80 metrics is usually referred as "PERCLOS").

AT89C51

Microcontroller

Driver Circuit

Vehicle Motor

LCD Display

Alarm

Steering wheel grip sensor

ADC 0808

MAX 232

Authentication

Processing

Recognition

Power supply

Figure 2: Eye tracking based driver fatigue detection and warning system

EYEMEAS (EM):the mean square percentage of the eyelid closure rating and the related technical brief from the Federal Highway Administrator that in order to detect the eyelid closures the face of test person was monitored and recorded, and rated the degree to which the driver’s eyes were closed from the moment to moment. The challenge related to the PERCLOS metrics is the automatic measurement of the eyelid position; however, successful attempts to measure eyelid position (and derive PERCLOS from it) are reported by Grace, where a camera monitors the face of driver. The PERCLOS metrics is measured directly and estimated with nonparametric methods for detecting drowsiness in drivers by Grace.

Image Processing

This approach analyzes the images captured by camera to detect physical changes of drivers, such as eyelid movement, eye gaze, yarn, and head nodding using MATLAB Software. For example, the PERCLOS system developed by W.W. Wirerwille used camera and image processing technique to measure the percentage of eyelid closure over the pupil over time. Although this video based system is non intrusive and will not cause annoyance to drivers. In addition this approach requires the camera to focus on a relative small area (around the driver’s eyes). It thus requires relative precise camera focus adjustment for every driver.

Steering wheel gripping pressure variability detection unit

As reported by Wylie et al, steering wheel variability is related to the amount of drowsiness in drivers (variability greater as driver become more drowsy) after being adjusted for road dependent effects. The technique which is used here is based on the fact that human body conducts current. Hence by using a conducting wire on non conducting steering wheel of Vehicle (as shown in Fig .3) and by using an Analog To Digital Convertor (ADC) and connected through a Transistor which act as a switch and when the driver hold the steering tightly more current flows through base of Transistor as parallel resistances made by our fingers add up in parallel and as a result net resistance decreases and base current increases. Hence this variation is converted by ADC into some threshold and whenever output is less than threshold it indicates Driver Drowsiness or Fatigue state.

One end of wire is connected to human body and human skin must be given +5v supply

ADC 0808

Microcontroller

+5v

Figure 3: Steering Wheel Gripping Pressure Variability Circuit

Output/ Warning Unit

For indication of warning we will use two approaches i.e. one by blowing alarm. Moreover the warning will be deactivated manually rather than automatically. So for this purpose a deactivation switch will be used to deactivate warning.

3. Result

The hardware developed "Driver Fatigue Detection Using Eye Tracking And Steering Wheel Unit System " is very advanced product related to driver safety in the roads as this product detects driver drowsiness and gives Warning in form of alarm and vibration in less than one second time which is the major achievement of this product.

Figure 4: Eyes are open

Figure 5: Eyes are closed

The Figure 4 and Figure 5 denote that cleared open and closed eye of driver. The image shown here is after the removal of noises. Noises are looks like some dots in image. Some commands are used for removing noises. After this, depending on closed eyes it gives intimation to driver in the form of alarm. If the driver is not get alert means then it makes the motor to get stop.

Figure 6: steering wheel gripping pressure

Figure 7: vehicle speed variation

Conclusion

This system will detect eye movement to detect the fatigue state of driver and gives warning in half second and also develop the performance record of driver in form of graph i.e. Speed versus Time and Steering gripping pressure variation which can be saved on hard disk for future reference. By monitoring the eyes using camera and using this new algorithm we can detect symptoms of driver fatigue early enough to avoid an accident. So this project will be helpful in detecting driver fatigue in advance and will gave a warning output in form of sound and vibration.