A-SLEEP

Embed Size (px)

Citation preview

  • 7/22/2019 A-SLEEP

    1/28

    Drowsiness Detection and Alerting Sensor System

    In partial fulfilment of the requirements in:MEREMEM

    940-1240, M, V507

    Submitted By:

    Ben Francis Alquiza

    Jorel Christopher Leonor

    RonnielOlfatoPatrick Roy Tan

    Jose Mari Zapanta

    Manufacturing Engineering and Management

    Submitted To:Dr. Nilo T. Bugtai

    April 15, 2011

  • 7/22/2019 A-SLEEP

    2/28

    Table of Contents

    Chapter 1Introduction .. 1

    1.1Introduction .. 11.2Background of the Study . 21.3Statement of the Problem . 31.4Significance of the Study . 41.5General Objective 41.6Specific Objectives .. 41.7Scope and Limitations

    1.7.1 Scope .. 51.7.2 Limitations . 5

    1.8Conceptual Framework 6

    Chapter 2Review of Related Literature ... 8

    2.1 Introduction .. 8

    2.2 Drowsy Driving and Automobile Crashes ... 82.2.1 Biology of Human Sleep and Sleepiness .. 9

    2.2.2 Sleepiness Impairs Performance . 10

    2.2.3 The Causes of Sleepy/Drowsy Driving ... 11

    2.2.3.1 Sleep Restriction or Loss . 112.2.3.2 Personal Demand and Lifestyle Choices . 11

    2.2.3.3 Sleep Fragmentation 12

    2.2.3.4 Circadian Factors . 122.2.4 Evaluating Sleepiness/Drowsiness .. 12

    2.2.4.1 Assessment for Chronic Sleepiness . 13

    2.2.4.2 Assessment for Acute Sleepiness . 14

    2.2.4.3 Vehicle-based Tools . 142.3 Drowsiness Detection System 14

    2.3.1 Blink Detection ... 15

    2.4 Driver Fatigue Detection Using Sensor Networks . 172.4.1 Action Unit Predictiveness . 18

    2.4.2 Embedded System ... 19

    2.5 Theoretical Framework .. 20

    References . 21

  • 7/22/2019 A-SLEEP

    3/28

    List of Figures:

    Figure 1.1 - trend on fatalities of the vehicular accidents in 2003 ........................................... 2

    Figure 1.1Conceptual Framework ... 6

    Figure 2.1 - Latency to sleep at 2-hour intervals across the 24-hour day .. 10

    Figure 2.2 - Performance slows with sleep deprivation . 11

    Figure 2.3Drowsiness Detection System ... 14

    Figure 2.4Blink Detection .. 15

    Figure 2.5 - Action Units ... 17

    Figure 2.6 - System Architecture ... 19

    Figure 2.7 - Theoretical Framework .. 19

    List of Tables

    Table 1.1 - Statistics of the causes of traffic accidents in the Philippines in 2003 .. 2

    http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsy.html#Figures%20(back)http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsy.html#Figures%20(back)
  • 7/22/2019 A-SLEEP

    4/28

    Definition of Terms

    Fatigue- a state of awareness describing a range of afflictions, usually associated with physical

    and/or mental weakness, though varying from a general state of lethargy to a specific work-

    induced burning sensation within one's muscles.

    Drowsiness - a state of near-sleep, a strong desire for sleep, or sleeping for unusually long

    periods

    Prototype- is an early sample or model built to test a concept or process or to act as a thing to

    be replicated or learned from.

    Circadian Cycle - is an endogenously driven roughly 24-hour cycle in biochemical,

    physiological, or behavioral processes.

    Vigilance - also termed sustained attention, is defined as the ability to maintain attention and

    alertness over prolonged periods of time.

    Pacemaker - is a medical device which uses electrical impulses, delivered by electrodescontacting the heart muscles, to regulate the beating of the heart.

    Chronic- is a disease that is long-lasting or recurrent. The term chronicdescribes the course ofthe disease, or its rate of onset and development. A chronic course is distinguished from a

    recurrent course; recurrent diseases relapse repeatedly, with periods of remission in between.

    Hurdle- is a moveable section of light fence. Traditionally they were made from wattle (wovensplit branches), but modern hurdles are often made of metal.

  • 7/22/2019 A-SLEEP

    5/28

    Chapter 1Introduction

    1.1 Introduction

    This study involves the design and fabrication of a prototype of a drowsiness detecting

    and alerting system to be used as a safety feature in automobile.

    This chapter will discuss the relevance of the study, its goals and objectives, the scope

    and finally its delimitations and limitations. Section 1.2 is about the background of the study.

    Section 1.3 states the statement of the problem. Section 1.4 presents the significance of the study.

    Section 1.5 states the general objective while the specific objectives are found in Section 1.6.

    Section 1.7 outlines the scope, delimitations and limitations of the study. Section 1.8 illustrates

    and explains the conceptual framework.

    1.2 Background of the Study

    Around the world, there is an estimate of 100,000 crashes related to driver fatigue.

    Almost 1,550 deaths, 71 thousand injured humans and around 12.5 billion dollars in economic

    losses are caused by driver fatigue. It is estimated that ten to twenty per cent of all fatal

    accidents and about five to ten per cent of all car accidents may be related to tired drivers. [1]

    In most cases in the Philippines, driver fatigue is a significant factor in a large number of

    vehicle accidents. In the Philippines, about one of every four traffic accidents is caused by

    drivers error. Among the drivers errors include that of drowsy driving. Graph (table1.1) shows

    the statistics of the causes of traffic accidents in the Philippines. Taking note on the trend on

    fatalities of the vehicular accidents (figure 1.1), the rise on such incidents from the past decade is

    alarming.

  • 7/22/2019 A-SLEEP

    6/28

    Table1.1 - Statistics of the causes of traffic Figure 1.1 - trend on fatalities of the

    accidents in the Philippines in 2003 vehicular accidents in 2003

    In the United States of America, the National Transportation Safety Board (NTSB)

    concluded that 52 per cent of 107 single-vehicle accidents involving heavy trucks were fatigue-

    related; in nearly 18 per cent of the cases, the driver admitted to falling asleep. Summarizing the

    US Department of Transportation's investigations into fatigue in the 1990s, the extent of fatigue-

    related fatal accidents is estimated to be around 30%. Research shows that driver fatigue is a

    significant factor in approximately 20% of commercial road transport crashes and over 50% of

    long haul drivers have fallen asleep at the wheel.

    In Australia, One study based on coronial and police reports found that fatigue played a

    part in only 5 per cent of fatal crashes in 1988. A more recent survey (for 1994) based on

    coronial and police reports found that fatigue played a part to about 18 per cent of fatal crashes.

    It included not only those crashes in which police identified fatigue as a cause, but also cases

    where the crash description suggested 'loss of concentration' had been a contributing factor. [2]

    Fatigue which causes drivers to fall asleep on the road has really become a worldwide

    problem. Drowsiness is also an issue that must be taken into consideration. Fatigue is different

    from drowsiness. In general, drowsiness is feeling the need to sleep, while fatigue is a lack of

  • 7/22/2019 A-SLEEP

    7/28

    energy and motivation. Drowsiness and apathy (a feeling of indifference or not caring about

    what happens) can be symptoms that go along with fatigue. [3]

    Keeping drivers awake is the main goal. If the drivers are awake and/or alerted, accidents

    can be prevented. We all know for a fact that alarm clocks can wake us up and heighten our

    senses. Relatively, if the study can apply this same principle, it can be used to wake up drowsy

    drivers, keep levels of alertness up and therefore reducing the probability of encountering a

    certain meeting with an accident or even death. This study will help save a lot of lives. When a

    sleeping or tired driver is involved in an accident, it is most likely that he has caused another

    motorist to be part or even the victim of his wrong doing. In some instances, the person

    responsible for the accident leaves the scene with minor injuries while the ones who just took

    part in the accident are the ones who are seriously injured.

    It is unfair but accidents are just a specific, unpredictable, unusual and unintended

    external action which occurs in a particular time and place, with no apparent and deliberate cause

    but with marked effects. It implies a generally negative outcome which may have been avoided

    or prevented had circumstances leading up to the accident been recognized, and acted upon, prior

    to its occurrence. Using this statement, the group can say that if the results of this study can be a

    good source of accident prevention, then accidents can be avoided by acting upon the issue

    before the inevitable occurs.

    1.3 Statement of the Problem

    There is a need to design and fabricate a system that is comprised of

    Electroencephalogram (EEG), Electrocardiogram (ECG), Electrooculography (EOG),

    Temperature and Face detection sensors in order to prevent sleeping related accidents in cars. It

  • 7/22/2019 A-SLEEP

    8/28

    will improve the accuracy of determining whether the car drivers are almost asleep or already

    under micro-sleep.

    1.4 Significance of the Study

    This aims to reduce car accidents caused by sleepy or drowsy drivers. Truck drivers and

    night shift workers can benefit a lot from the system. Before any accident can occur, the system

    can help prevent it. The system could help a lot in lessening the occurrences of car accidents

    caused by drowsy driving and help save tons of lives. If this system is implemented and to be

    made a standard feature in all vehicles, drowsy driving can be prevented and in the long run,

    eliminated from the charts as a cause of vehicular accidents.

    The system could also be applied in medical fields. For example, it can be used as an

    additional system in helping to determine whether a patient is really asleep whenever they

    undergo surgery.

    1.5 General Objective

    To design and fabricate a prototype of a drowsiness detecting and alerting sensor systemto be used as a safety feature in cars that would be able to sense and identify if the driver

    of the car is about to fall asleep and then wake up the driver to prevent accidents.

    1.6 Specific Objectives

    To use measure and interpret the brain waves using an EEG machine and sense whetherthe driver is drowsy or asleep.

  • 7/22/2019 A-SLEEP

    9/28

    To track and analyse the movement of the eyes with the use of an EOG machine whichcould be the basis on concluding if the driver is sleepy or not.

    To determine if the driver is drowsy or asleep by using an ECG machine to verify thedrivers condition through reading the data produced by his heart rate.

    To measure the distance between the pupils in order to quantify if the driver is about tosleep by using the facial recognition system.

    To use the temperature of the driver to conclude whether the driver is sleeping or not. To alert and wake up the driver when he is about to sleep without startling them.

    1.7 Scope and Limitations

    1.7.1 Scope

    The study is concerned with the detecting of drowsiness and alerting of the driver only It is concerned with the identification of where to locate sensors in an automobile only. It takes into consideration the time or part of the day the driver is driving.

    1.7.2 Limitations

    Abnormalities, illnesses and ailments (e.g. cross-eyed, cold, cough, asthma, high bloodpressure, etc.) will not be taken into consideration.

    The study does not cover the detecting and/or determining whether the driver is under theinfluence of alcohol and/or drugs that affects the physiological state of the driver.

    It does not cover how to attach and/or to set up the sensor in an automobile.

  • 7/22/2019 A-SLEEP

    10/28

    1.8 Conceptual Framework

    The proposed system would allow the interpretation of input data, gathered from the different

    sensors, to the control unit. The control unit is preferably run by a neural network algorithm that

    allows the input data to be trained and tested to a more accurate and precise output. The model

    should be able to recognize the level of drowsiness. Certain factors are to be considered (i.e

    microcontrollers, amplifiers, filters) in creating the control unit. The model should be able to

    accomplish the following tasks:

    For the face detection system, it should be able to evaluate the level of position of theeyes with respect to its original position and its state at micro sleep with respect to time

    For the EEG, it should be able to evaluate the micro sleep and brain activity whetheractive or not

    For the ECG, it should be able to measure the heart pulse rate if it is in the range offatigue

    For the EOG, it should be able to determine the activity of the eye whether the driver isdrowsy.

    For the thermal system, it should be able to recognize the core body temperature and theshell temperature if it is in the state of drowsiness

  • 7/22/2019 A-SLEEP

    11/28

    Figure 1.2Conceptual Framework

  • 7/22/2019 A-SLEEP

    12/28

    Chapter IIReview of Related Literature

    2.1 Introduction

    A lot of studies have been conducted regarding sleeping or experiencing drowsiness

    while driving. This chapter will discuss about the related literature regarding the study. Section

    2.2 shows an overview on drowsy driving and automobile crashes. Section 2.2.1 discusses the

    biology of human sleep and sleepiness while section 2.2.2 discusses the possible effects of

    sleepiness to the performance of the driver. Section 2.2.3 explains the causes of drowsiness in

    driving. Section 2.2.4 enumerates the different and/or possible measuring and evaluating systems

    for sleepiness. Section 2.3 and 2.4 proposes the most feasible alerting methods that are available

    today. Section 2.5 discusses about the theoretical framework.

    2.2 Drowsy Driving and Automobile Crashes

    National Heart, Lung, and Blood Institute

    National Center on Sleep Disorders Research

    The article shows the basic elements of the state of sleepiness/drowsiness, including the

    factors that affect them. In the state of sleepiness/drowsiness, there is a direct correlation

    between drowsiness and accidents. Given the statistics in the Philippines, the trend is alarming in

    the past decade and certain measures are needed to analyze this trend for further improvement on

    safety.

  • 7/22/2019 A-SLEEP

    13/28

    2.2.1 Biology of Human Sleep and Sleepiness

    Sleep is a neurobiological need with predictable patterns of sleepiness and wakefulness.

    Sleepiness or drowsiness results from the sleep component of the circadian cycle (body clock) of

    sleep and wakefulness, restriction of sleep, and/or interruption or fragmentation of sleep.

    The terms "fatigue" and "inattention" are sometimes used interchangeably with

    sleepiness; however, these terms have individual meanings [5]. Strictly speaking, fatigue is the

    consequence of physical labor or a prolonged experience and is defined as a disinclination to

    continue the task at hand. In regard to driving, a psychologically based conflict occurs between

    the disinclination to drive and the need to drive. One result can be a progressive withdrawal of

    attention to the tasks required for safe driving. Inattention can result from fatigue, but the crash

    literature also identifies preoccupation, distractions inside the vehicle, and other behaviors as

    inattention [6].

    Figure 2.1. Latency to sleep at 2-hour intervals across the 24-hour day.Testing during the daytime followed standard Multiple Sleep Latency Test

    procedures. During the night, from 2330 to 0800 hours (based on a 24 -hour clock), subjects were awakened every 2 hours for 15 minutes, and

    latency of return to sleep was measured. Elderly subjects (n = 10) were 60 to 83 years of age; young subjects (n = 8) were 19 to 23 years of age

    (Carskadon and Dement, 1987).

    http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsy.html#Figures%20(back)http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsy.html#Figures%20(back)http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsy.html#Figures%20(back)
  • 7/22/2019 A-SLEEP

    14/28

    Figure 2.2 Performance slows with sleep deprivation. A summary of data (Kribbs, Dinges, 1994) on reaction to an event marker presented to a

    subject every 4 seconds or so over a 10-minute period. As reaction time is longer, the inverse value is reduced, indicating a slowing of the

    perception/reaction response. The response to an event marker slows more across time in the sleep-deprived (very sleepy) subject than in a

    subject who has had normal amounts of sleep.

    Sleepiness causes automobile crashes because it impairs performance and can ultimately

    lead to the inability to resist falling asleep at the wheel. Critical aspects of driving impairment

    associated with sleepiness are reaction time, vigilance, attention, and information processing.

    2.2.2 Sleepiness Impairs Performance

    Sleepiness leads to crashes because it impairs elements of human performance that are

    critical to safe driving [7]. Relevant impairments identified in laboratory and in vehicle studies

    include:

    Slower reaction time. Sleepiness reduces optimum reaction times, and moderately sleepypeople can have a performance impairing increase in reaction time that will hinder

  • 7/22/2019 A-SLEEP

    15/28

    stopping in time to avoid a collision (Dinges, 1995). Even small decrements in reaction

    time can have a profound effect on crash risk, particularly at high speeds.

    Reduced vigilance. Performance on attention based tasks declines with sleepiness,including increased periods of non-responding or delayed response [7]

    Deficits in information processing. Processing and integrating information takes longer,the accuracy of short-term memory decreases, and performance declines [7]

    2.2.3 The Causes of Sleepy/Drowsy Driving

    Although alcohol and some medications can independently induce sleepiness, the

    primary causes of sleepiness and drowsy driving in people without sleep disorders are sleep

    restriction and sleep fragmentation.

    2.2.3.1 Sleep restriction or loss.

    Both external and internal factors can lead to a restriction in the time available for sleep.

    External factors, some beyond the individuals control, include work hours, job and family

    responsibilities, and school bus or school opening times. Internal or personal factors sometimes

    are involuntary, such as a medication effect that interrupts sleep. Often, however, reasons for

    sleep restriction represent a lifestyle choicesleeping less to have more time to work, study,

    socialize, or engage in other activities.

    2.2.3.2 Personal Demands and Lifestyle Choices

    Different lifestyles of different persons may cause drowsiness. It is because the amount of

    effort and energy they put up into it that makes a difference in their level of drowsiness. For

  • 7/22/2019 A-SLEEP

    16/28

    example, an athletes level of drowsiness, given the amount of energy they put into their wo rk, is

    different from that of a person who is relaxing. This lifestyle is a big factor that affects

    drowsiness or sleepiness.

    2.2.3.3 Sleep fragmentation

    Sleep is an active process, and adequate time in bed does not mean that adequate sleep

    has been obtained. Sleep disruption and fragmentation cause inadequate sleep and can negatively

    affect functioning [7]. Similar to sleep restriction, sleep fragmentation can have internal and

    external causes. The primary internal cause is illness, including untreated sleep disorders.

    Externally, disturbances such as noise, children, activity and lights, a restless spouse, or job-

    related duties (e.g., workers who are on call) can interrupt and reduce the quality and quantity of

    sleep. Studies of commercial vehicle drivers present similar findings.

    2.2.3.4 Circadian factors

    The circadian pacemaker regularly produces feelings of sleepiness during the afternoon

    and evening, even among people who are not sleep-deprived [7]. Shift work can also disturb

    sleep by interfering with circadian sleep patterns.

    2.2.4 Evaluating Sleepiness/Drowsiness

    Possible techniques for detecting drowsiness in drivers can be generally divided into the

    following categories: sensing of physiological characteristics, sensing of driver operation,

    sensing of vehicle response, monitoring the response of driver.

  • 7/22/2019 A-SLEEP

    17/28

    An ideal measure of sleepiness would be a physiologically based screening tool that is

    rapid and suitable for repeated administration [8]. A measuring system would be performance

    based and in vehicle, linked to alerting devices designed to prevent the driver from falling asleep.

    The current tools for the assessment of sleepiness are based on questionnaires and

    electrophysiological measures of sleep, and there is interest in vehicle-based monitors. A

    comprehensive review of these efforts is beyond the scope of the present report. In the following

    brief discussion, some tools for the assessment of sleepiness are described to illustrate the

    different subjective and objective measures of chronic and situational (acute) sleepiness and the

    vehicle-based technology to sense sleepiness.

    2.2.4.1 Assessment for chronic sleepiness

    The Epworth Sleepiness Scale (ESS) [10] is an eight-item, self-report measure that

    quantifies individuals sleepiness by their tendency to fall asleep. Laboratory tools for measuring

    sleepiness include the Multiple Sleep Latency Test (MSLT) [12] and the Maintenance of

    Wakefulness Test (MWT) [8]. The MSLT measures the tendency to fall asleep in a standardized

    sleep-promoting situation during four or five 20- minute nap opportunities that are spaced 2

    hours apart throughout the day and in which the individual is instructed to try to fall asleep.

    Sleep is determined by predefined brain wave sleep-staging criteria. In the MWT,

    individuals are instructed to remain awake, and the time it takes (if ever) in 20 minutes to fall

    asleep by brain wave criteria is the measure of sleepiness. The MSLT and MWT were developed

    for neurophysiologic assessment and are sensitive to acute as well as chronic sleep loss.

  • 7/22/2019 A-SLEEP

    18/28

    2.2.4.2 Assessment for acute sleepiness

    Acute sleepiness is defined as a need for sleep that is present at a particular point in time.

    The Stanford Sleepiness Scale (SSS) [12] is an instrument that contains seven statements through

    whichpeople rate their current level of alertness (e.g., 1= feeling...wide awake to 7= ...sleep

    onset soon...). The scale correlates with standard performance measures, is sensitive to sleep

    loss, and can be administered repeatedly throughout a 24-hour period. In some situations, the

    scale does not appear to correlate well with behavioral indicators of sleepiness; in other words,

    people with obvious signs of sleepiness have chosen ratings 1 or 2.

    2.2.4.3 Vehicle-based tools

    There are some in vehicle systems that are intended to measure sleepiness or some

    behavior associated with sleepiness in commercial and non-commercial driving. Examples

    include brain wave monitors, eye-closure monitors, devices that detect steering variance and

    tracking devices that detect lane drift [7].

    2.3 Drowsiness Detection System

    by Kenji Ogawa and MitsuoShimotani*

    The article describes a system that detects when drivers are becoming drowsy and sound

    a warning promise to be a valuable aid in preventing accidents. This system was developed and

    innovated at Mitsubishi Electric. The system analyzes facial images of the driver to determine

    blinking behavior, which it uses as a measure of driver alertness [17]. An accurate real-time

    drowsiness detector prototype was manufactured and tested in actual vehicles.

  • 7/22/2019 A-SLEEP

    19/28

    Figure 2.3Drowsiness Detection System

    2.3.1 Blink Detection

    A small CCD camera mounted in the vehicle instrument panel was used to capture

    images of the subjects face while driving [17]. Several hurdles and problems were noticed in

    developing an effective method to measure blink duration.

  • 7/22/2019 A-SLEEP

    20/28

    Figure 2.4Blink Detection

    There is a distinct relationship between alertness and blink duration. Fig. 2 shows the

    relationship between decreasing alertness and blink duration. For each minute, the figure shows a

    histogram for various blink durations, reaction times and subjective drowsiness evaluations.

    Long blink durations became more frequent as the subjects began to report feeling drowsy.

    The average blink duration for alert subjects varies with the individual. Long blink

    durations of a half-second or more correspond to subjective evaluations of slightly or moderately

    sleepy. The one subject who remained completely alert recorded no blink durations over a half-

  • 7/22/2019 A-SLEEP

    21/28

    second. Blink durations of a half-second or longer were always present when reaction times over

    two seconds were recorded.

    2.4 Driver Fatigue Detection Using Sensor Networks

    Mr. Swapnil V. Deshmukh #1, Ms.Dipeeka P. Radake*2, Mr. Kapil N. Hande#3

    In the field of automotive safety research, a method to monitor and to detect a

    fatigue/drowsy or a drunken driver has been studied for many years. Previous research uses

    sensors such as an infrared camera for pupil detection or voice to detect fatigue, or image

    processing to detect drivers expression. These approaches are able to detect drivers fatigue;

    however, these methods are not driver adaptable nor interactive with the outside driving

    situation. We are proposing a drivers fatigue approach for real-time detection.

    The system consists of a sensors directly pointed towards the drivers face. The input to

    the system is a continuous stream of signals from the sensors. The system monitors the drivers

    eyes to detect micro-sleeps (short periods of sleep lasting 3 to 4 seconds), monitors the drive rs

    jaw to detect jaw movement and monitors to detect driver pulse from finger using LED & LDR

    assembling. The system can analyze the eyes lid movement, jaw movement, variation in pulse

    rate from the driver compute it as well as compare signal [15].

    Technological approaches for detecting and monitoring fatigue levels of the driver

    continue to emerge and many are now in the development, validation testing, or early

    implementation stages [13]. Previous studies have reviewed available fatigue detection and

  • 7/22/2019 A-SLEEP

    22/28

    prediction technologies and methodologies. A method using an infrared camera to monitor pupil

    of the eye is the most popular method. Other research uses EEG to monitor driver fatigue [4].

    2.4.1 Action Unit Predictiveness

    In order to understand the action unit predictiveness in drowsiness, MLR was trained on

    frame wise outputs of each facial action individually. Examination of the A for each action unit

    reveals the degree to which each facial movement is associated with drowsiness in this study.

    The As for the drowsy and alert states are shown in Figure 2.5. Performance was evaluated in

    terms of the area. For all of the novel subject analysis, the output for each feature was summed

    over a temporal window of seconds before computing. Cross validation was performed with

    trained on subjects and tested on subject at a time [13].

    Figure 2.5 - Action Units

    An active safety system needs to be developed to reduce number of automobiles

    accidents due to reduced vigilance. Drowsiness in drivers can be generally divided into the

  • 7/22/2019 A-SLEEP

    23/28

    following categories: sensing of physiological characteristics, sensing of driver operation,

    sensing of vehicle response, monitoring the response of driver.

    Among these methods, the techniques based on human physiological phenomena are the

    most accurate. This technique is implemented in two ways: measuring changes in physiological

    signals, such as brain waves, heart rate, and eye blinking; and measuring physical changes such

    as sagging posture, leaning of the drivers head and the open/closed states of the eyes.

    2.4.2 Embedded System

    The embedded system will be used to improve the accuracy of fatigue detection as

    compared to existing fatigue detection system. Certain objectives of the system include the

    following: measure eye blinks, heart rate, jaw movement; volume of auditory output should

    range from 50dB to 90dB [13].

    Systems Architecture will consist of three essential components as shown in Figure 2.6,

    which is shown below. This is the basic outline of the system. Further adjustments may be

    necessary to achieve optimal output

    INPUT UNIT:Sensor can sense the behavior of user.

    CONTROL UNIT:For calculating, decision making and evaluation

    DISPLAY UNIT / OUTPUT UNIT:It is used to display the result in various forms (i.e. sound, vibration, visual stimulus)

    [16].

  • 7/22/2019 A-SLEEP

    24/28

    Figure 2.6 - System Architecture

    2.5 EOG-based drowsiness detection

    Antoine Picot,Sylvie Charbonnier,Alice Caplier

    This paper presents an original drowsiness detection method based on the fuzzy merging

    of several eye blinking features extracted from an electrooculogram (EOG). These features are

    computed each second using a sliding window. This method is compared to two supervised

    learning classifiers: a prototype nearest neighbours and a multilayer perceptron.

    The method proposed reaches very good performances with 82% of true detections and

    13% of false alarms. The best results obtained by the supervised learning classification methods

    are only 72% of true detection and 26% of false alarms, which is far worse than the fuzzy

    method [17]. It is shown that the fuzzy method overtakes the other methods because it is able to

    take into account the fact that drowsiness symptoms occur simultaneously and in a repetitive way

    on the different features during the epoch to classify, which is of importance in the drowsiness

    decision-making process.

    http://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Picot&v_1=Antoinehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Charbonnier&v_1=Sylviehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Caplier&v_1=Alicehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Caplier&v_1=Alicehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Caplier&v_1=Alicehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Charbonnier&v_1=Sylviehttp://hal.archives-ouvertes.fr/index.php?action_todo=search&s_type=advanced&submit=1&search_without_file=YES&f_0=LASTNAME&p_0=is_exactly&f_1=FIRSTNAME&p_1=is_exactly&l_0=and&halsid=atdpn067t71kq8hlrt732s76n1&v_0=Picot&v_1=Antoine
  • 7/22/2019 A-SLEEP

    25/28

    2.6 Theoretical Framework

    The certain studies, shown in figure 2.7, contribute largely on the said system. Many

    theoretical aspects conducted on those studies will be integrated on the groups concept. This

    would allow achieving an optimal framework for the group to base on.

    Figure 2.7 - Theoretical Framework

    Evaluation of Drowsiness During Driving using

    Electrocardiogram - A Driving Simulation Study

    Masaru Tasaki Motoaki Sakai Mai Watanabe Hui Wang Daming

    Wei

    Published in: Proceeding CIT '10 Proceedings of the 2010 10th

    IEEE International Conference on Computer and Information

    Technology IEEE Computer Society Washington, DC, USA

    Driver fatigue Detection Using Sensor Network

    Mr. Swapnil V. Deshmukh #1, Ms.Dipeeka P. Radake*2, Mr.

    Kapil N. Hande#3

    #Department of Computer Science & Engineering, NYSSCER,Nagpur, India

    EOG-based drowsiness detection: Comparison

    between a fuzzy system and two supervised

    learning classifiers

    Antoine Picot 1, Sylvie Charbonnier 1,

    Alice Caplier 1

    Grenoble Images Parole Signal Automatique

    Universit Joseph Polytechnique de Grenoble

    Drowsiness detection system

    United States Patent 6822573

    Otman Basir, Jean Bhavani, Fakhreddine Karay, Kristopher

    Desrochers

    Drowsiness

    Detection and

    Alerting System

  • 7/22/2019 A-SLEEP

    26/28

    AppendixEconomic Analysis

    Statistics in the US

    According to the National Sleep Foundation's 2002 "Sleep in America" poll, about one-half of

    adult drivers (about 100 million people) say they've driven a vehicle in the past year while

    feeling drowsy. Almost two in 10 people (about 32 million) have actually fallen asleep at the

    wheel. One percent (approximately two million drivers) had an accident because they dozed off

    or were too tired to drive.

    Statistics in the Philippines

    In most cases in the Philippines, driver fatigue is a significant factor in a large number of

    vehicle accidents. In the Philippines, about one of every four traffic accidents is caused by

    drivers error. Among the drivers errors include that of drowsy driving. Graph (figure 2.1)

    shows the statistics of the causes of traffic accidents in the Philippines. Taking note on the trend

    on fatalities of the vehicular accidents (figure 2.2), the rise on such incidents from the past

    decade is alarming.

    Table1.1 - Statistics of the causes of traffic Figure 1.1 - trend on fatalities of the

    accidents in the Philippines in 2003 vehicular accidents in 2003

  • 7/22/2019 A-SLEEP

    27/28

    References:

    [1] http://www.smartmotorist.com/traffic-and-safety-guideline/driver-fatigue-is-an-

    important-cause-of-road-crashes.html

    [2] http://www.smartmotorist.com/traffic-and-safety-guideline/driver-fatigue-is-an-

    important-cause-of-road-crashes.html

    [3] http://www.nlm.nih.gov/medlineplus/ency/article/003088.htm

    [4] Learn to sleep well" (Duncan Baird, 2000) Niedermeyer E. and da Silva F.L.

    (2004).Electroencephalography: Basic Principles, Clinical Applications, and Related

    Fields. Lippincot, Williams & Wilkins. Retrieved from:

    http://www.nhtsa.gov/people/injury/drowsy_driving1/drowsyref.html

    [5] Treat et al: Tri-level study of the causes of traffic accidents: final report. Volume I:

    Causal factor tabulations and assessments. Institute for Research in Public Safety, Indiana

    University; 1979. DOT Publication No.: DOT HS-805085.

    [6] Dinges D, Kribbs N: Performing while sleepy: effects of experimentally-induced

    sleepiness. In: Monk T, editor. Sleep, sleepiness, and performance. New York: John

    Wiley & Sons; 1991. pp. 98-128.

    [7] Mitler M et al.: Maintenance of wakefulness test: a polysomnographic technique for

    evaluating treatment efficacy in patients with excessive somnolence.

    ElectroencephalogrClinNeurophysiol 1982;53:658-61

    [8] Pack A et al.: Characteristics of crashes attributed to the driver having fallen asleep

    Accid Anal Prev 1995;27(6):769-75

    [9] http://ezinearticles.com/?Sleep-and-Body-Temperature---The-Connection&id=177405

    [10] Nerad, J. Falling asleep at the wheel. 2006.

  • 7/22/2019 A-SLEEP

    28/28

    [11] THE INTERNATIONAL CLASSIFICATION OF SLEEP DISORDERS, REVISED,

    American Academy of Sleep Medicine, 2001

    [12] http://www.ijest.info/docs/IJEST-NCICT-020-190.pdf

    [13] Rumble strips: A wake up call for drowsy drivers. Retrieved fromhttp://www.usroads.com/journals/rmej/0002/rm000201.htm

    [14] Squatriglia, C. Vibrating seat warns of imminent danger. 2010.

    [15] Vibrating motors in automotive applications. Retrieved from

    http://www.precisionmicrodrives.com/vibration-motor/vibration-motor-

    markets/automotive

    [16] Bulling, A. Eye movement analysis for activity recognition using electrooculography.

    [17] EOG-based drowsiness detection: Comparison between a fuzzy system and two supervised

    learning classifiers