Ngo Duy Kien _Thesis Version 10_25

download Ngo Duy Kien _Thesis Version 10_25

of 52

Transcript of Ngo Duy Kien _Thesis Version 10_25

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    1/52

    Error: Reference source not foundVIETNAM NATIONAL UNIVERSITY, HA NOI

    UNIVERSITY OF ENGINEERING AND TECHNOLOGY

    --------

    Ngo Duy Kien

    FALL DETECTION BASED ON

    ACCELEROMETER SENSOR

    GRADUATION THESIS

    Major Field: Computer Science

    Ha Noi 2012

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    2/52

    VIETNAM NATIONAL UNIVERSITY, HA NOI

    UNIVERSITY OF ENGINEERING AND TECHNOLOGY

    --------

    Ngo Duy Kien

    FALL DETECTION BASED ON

    ACCELEROMETER SENSOR

    GRADUATION THESIS

    Major Field: Computer Science

    Supervisors: Assoc. Prof. Bui The Duy

    Co-Supervisor: PhD. Vu Thi Hong Nhan

    Ha Noi 2012

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    3/52

    AUTHORSHIP

    I hereby declare that the work contained in this thesis is of my own and has not been previously

    submitted for a degree or diploma at this or any other higher education institution. To the best of my

    knowledge and belief, the thesis contains no materials previously published or written by another

    person except where due reference or acknowledgement is made.

    Signature:

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    4/52

    SUPERVISORS APPROVAL

    I hereby approve that the thesis in its current form is ready for committee examination as arequirement for the Bachelor of Computer Science degree at the University of Engineering and

    Technology.

    Signature:

    iii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    5/52

    ACKNOWLEDGEMENT

    First of all, I wish to express my respect and my deepest thanks to my advisers

    Assoc.Prof. Bui The Duy and PhD. Vu Thi Hong Nhan, University of Engineering and

    Technology, Viet Nam National University, Ha Noi for their enthusiastic guidance, warm

    encouragement and useful research experiences.

    I would like to gratefully thank all the teachers of University of Engineering and

    Technology, VNU for their invaluable knowledge which they gave to me during four

    academic years.

    I would also like to say my special thanks to my friends in K53CA class, University of

    Engineering and Technology, VNU, especially Tran Nguyen Le in the Behavior

    Recognition group for their helpful discussions.

    Last, but not least, my family is really the biggest motivation behind me. My parents

    and my brother always encourage me when I have stress and difficulty. I would like to send

    them my gratefulness and love.

    Ha Noi, May 24th

    , 2012Ngo Duy Kien

    iv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    6/52

    FALL DETECTION BASED ON ACCELEROMETER SENSOR

    Ngo Duy Kien

    QH-2008-I/CQ, Computer Science

    Abstract:

    Nowadays, the elderly population is rapidly growing. Fall-related injuries are a centralissue for this population. Falls account for approximately half of all admissions- related injuries.

    Falls cause not only broken bones and other health injuries, but also cause psychological trauma

    which can reduce the independence and confidence in communication. Techniques for detecting

    and tracking down the movement are instrumental in dealing with this problem.

    Elderly people often want to live at home, therefore, new technologies need to be able to

    support automatic fall detection, which guarantee their living independence and security. This

    thesis meets the challenge of different types of motions as part of a system designed to fulfill the

    demand for wearable device to collect data for fall and non-fall analysis. this study focuses on

    evaluating different low-complexity fall detection algorithms by using tri-axial accelerometer

    sensor. Towards this task, a scenario in which both fall and non-fall events in different situations is

    set up. Some types of intentional falls would be standardized for the fall data. They include:

    forward falls, backward falls and lateral falls right and lateral falls left, performed with legs straight

    and flexed in fire middle aged subjects. Data from non-fall (activity of daily living) were used as

    the reference group. Some approaches with increasing detection complexity are investigated andemployed for the real time classification in the characteristics of a fall event: beginning of the fall,

    falling velocity, variance of the fall, the accelerometer inclination angles, the acceleration vector

    changes and fall impact after the fall. The comparison of classification algorithms is made by a

    practical machine learning tool WEKA. We also adapt for different gaits of different people.

    The experimental results indicate that fall detection using a tri-axial accelerometer sensor is

    efficient, and can reach a sensitivity of 94%.

    Keyword: fall detection, accelerometer, activity of daily living

    v

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    7/52

    Table of Contents...........................................................................................................vi

    List of Figures...............................................................................................................viiList of Table.................................................................................................................viii

    Introduction....................................................................................................................ix

    Related Work..................................................................................................................xi

    Scenarios of falls in Our Daily Activities..................................................................xxiii

    Fall Detection Process.............................................................................................xxviii

    Experimental Setup and evaluation........................................................................xxxvii

    Chapter 6......................................................................................................................xlii

    Conclusion and Future work......................................................................................................................................................xlii

    Bibliography..................................................................................................................44

    Appendix A...................................................................................................................47

    .......................................................................................................................................48

    vi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    8/52

    Figure 2-1: Reading trunk pitches rolocity V v and horizontal velocity Vh. These

    velocities were obtained without markers from the 3D trajectory of a person who brutally sitsdown and falls. The different action are: the person(a) stands up,(b)sits down, (c) is seated,(d)stands up again, (e)remains stand up and (f) falls...................................................................xiii

    Figure 2-2: Acceleromter Sensor.....................................................................................xiv

    Figure 2-3: Reading trunk pitches and roll gyroscope....................................................xvi

    Figure 2-4: Overlapping and non-overlapping sampe Fall and ADL data......................xvi

    Figure 2-5: Bi-axial gyroscope fall-detection algorithm flow chart..............................xvii

    Figure 2-6: Fall detection algorithm operation example for upper and lower fallthreshold, using an artificial example signal..............................................................................xviii

    Figure 3-7: The stage of fall...........................................................................................xxiv

    Figure 3-8: The four phases of a fall event....................................................................xxvi

    Figure 4-9: The general model cycle.............................................................................xxix

    Figure 4-10: fall annotation for a single fall...................................................................xxx

    Figure 4-11: Low-pas and high-pass filter algorithm..................................................xxxii

    Figure 4-12: RSS plot of a forward fall.......................................................................xxxiv

    Figure 4-13: Accelerometer inclination angle.............................................................xxxvi

    Figure 4-14: Fall Detection Flow Chart.......................................................................xxxvi

    vii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    9/52

    Table 4.1: sample data acquired from acceleromter.......................................................xxx

    Table 5.2: Events Sequence in the test Scenario.......................................................xxxviii

    Table 5.3: Effect of body profile on the performance.......................................................xl

    Table 5.4: The samples for each category........................................................................xli

    Table 5.5: The test result...................................................................................................xli

    viii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    10/52

    Chapter 1

    In the modern society, the proportion of elderly people is increasing rapidly. The

    ratio of elderly people living alone in the house is quite high. Fall-related injury is a

    central issue of elderly people, both home environment and hospital which affects the

    overall quality of life either at home or hospital. They always want to live in a place

    which ensures the independence and safety in life. Most of falls occur when walking,

    standing or sitting down, or trying to find something. Because of this, hence, newtechnologies to guarantee and assist their lives are absolutely necessary. This is a factor

    to promote the development of technology. In the modern society, smart phone

    applications are developing faster and faster. Therefore, with a phone equipped sensors,

    researchers have developed applications for fall detection based on accelerometer

    sensor. These applications are mainly developed based on the threshold algorithm,some works involve the use of machine learning techniques for the detection of fall and

    movement classification

    In this thesis, we investigate some typical machine-learning methods for activity

    recognition and especially for fall recognition. We focus on evaluating and choosing

    which characteristics and machine learning algorithm would maximize the recognition

    accuracy. Some types of activities which can be recognized are sitting, running, lying or

    falling. However, most people usually think that falling recognition is the most

    ix

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    11/52

    important in our lives. Fall detection has become a major health concern recently. Falls

    account for approximately half of all admissions which are related to injury. Falls cause

    not only disabling fractures and other health injuries but also traumas which can reduce

    the independence and confidence when communicating.

    There are many approaches that can be used to construct a fall detection systems.

    The first one is based on image and video system and the second one is based on

    accelerometer sensor. The later is our concern in this study because with the rapid

    advances in wireless network and tiny sensors they can be embedded easily in the

    mobile devices which can be carried anywhere. This enables the elders status to be

    monitored anytime and anywhere and aided when necessary. To evaluate the

    performance of classification methods for the application of fall detection, we set up a

    scenario in which the test subject does some different daily activities from which dataare collected. From these datasets, we performed computing the attribute for

    classification, namely length of the acceleration vector, the variance of the length, speed

    of change in acceleration along the z-axis , the acceleration vector changes, and the

    accelerometer inclination angles. with these study result , based on experience, we use

    SVM fucntion for classification which is supported by the software toolkit Weka. The

    thesis is structured as follows:

    Chapter 2 presents a detailed overview about of related work on recognitionactivities and fall detection.

    Chapter 3 gives a knowledge background in fall

    Chapter 4 describes the design of a simulation system for fall detection in detail.

    Firstly, in order to get data to train the classifier, we design a number of dataset

    collection forms to extract features and assign them labels automatically.

    Chapter 5, we presents an experiment to emphasize the importance of feature

    extraction in fall detection. In addition, we described the construction ofaccelerometer sensor system and evaluate its performance using support vector

    machine and other algorithms.

    Chapter 6 summarizes the thesis some conclusions and future work directions.

    x

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    12/52

    Chapter 2

    Accelerometer and gyroscopes are the two most popular methods in most of the

    research on fall. Threshold for acceleration, the change of velocity and angles were

    usually applied in some researches. By applying a simple threshold to the acceleration

    and using a tri-axial accelerometer worn on the chest, Lee, Nguyen, Cho (2009)

    detected falls with 98. Bourke and Lyons (2008) usedbiaxial gyroscope worn on the

    chest, they applied threshold to the peaks in the angular velocity, angular acceleration

    and the change of angle.

    In addition to applying threshold for fall detection, machine-learning algorithm

    seems to be a good method instead of threshold algorithm algorithm. One example for

    using machine learning is Shan (2010) and Yuan and Zhang (2006), both studies which

    used tri-axial accelerometer wornon the waist. They used SVM algorithm to classify

    various features from accelerometer sensor data.

    Especially, Particular interest to us is research of Lie (2009). Although their

    result of fall detection is lower than previously research, it may be due to the more

    experimental data. They apply threshold, angles and velocities. Based on this, potential

    fall and the activity will be detected after fall. Their methods sometimes do not lie down

    quickly. And these situations will be addressed by the classify accelerometer sensor on

    this thesis.

    xi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    13/52

    2.1. Fall Detection using 3D Head Trajectory Extracted

    from a Signal Camera Video Sequence

    Falls are one of the greatest dangers for elderly people living alone. They

    may be unconscious after suffering the fall. Computer vision system provides an

    automatic solution to overcome the limitations of researchers. Some research has

    developed fall detection system by using the image sensor. One example is the work of

    Lee and Mihailidis who detect falls by using a camera mounted on the ceiling determine

    the specific silhouette 5.2. This is a new method using 3D data of the head trajectory for

    monitoring the movement of the fall.

    The method of this research will be based on three steps:

    Head tracking: because of the fact that the head can be often seen in the pictureand there is a large movement during a fall

    3D tracking: the head is monitored with a particle filter to extract a 3D

    trajectory

    Fall Detecting: a fall is detected by using 3D velocities which are computed

    from the 3D trajectory of the head.

    xii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    14/52

    Figure 2-1: Reading trunk pitches rolocity V v and horizontal velocity Vh. These

    velocities were obtained without markers from the 3D trajectory of a person who

    brutally sits down and falls. The different action are: the person(a) stands up,

    (b)sits down, (c) is seated, (d)stands up again, (e)remains stand up and (f) falls

    This method required a set of video in various situations such falls and normal

    situations like sitting down or squatting. Based on the characteristics of the image

    sequences, the result will be showed after analyzing with OpenCV library (Intel Open

    Source Computer Vision Library)

    The result obtained in this research is quite high. However, in some cases it

    depends on the location and specific circumstance. Camera is not always everywhere.

    Moreover, using camera usually makes the user feel unnatural and lose freedom in the

    daily activities. It affects the privacy of individuals. Consequently, using acceleration

    sensor for fall detection is an effective approach to ensure personal information can be

    kept secure and applied in every situation.

    xiii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    15/52

    2.2. Fall Detection using accelerometer sensor

    Accelerometer usually provides the acceleration readings in direction of x, y, z

    axis. Accelerations are these directions are presented by xA , yA , zA respectively. As

    illustrated in Figure 2-2

    Figure 2-2: Acceleromter Sensor

    X-axis has positive direction toward the right side of the device

    Y-axis has positive direction toward the top of the device

    Z-axis has positive direction toward the front of the device

    Noury and Rumeauindicate two approaches to detect fall in their research 5.2. The

    first place, which is the more common approach, is an analytical method and the second

    place is with machine learning techniques. An example for machine learning approach is

    the research of Mitja and Bostjan 5.2. This system is a visual basic system. By using

    markers, they tagged different point in body of users and used it as reference points.

    They found that the angle and the reference points between them is very reliable data

    source for extracting features. With other machine-learning algorithm, Support Vector

    machine (SVM) was the best choice for research and followed by Random Forest.

    In our research for fall detection is mainly based on the machine learning

    techniques and some use analytical models.

    xiv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    16/52

    2.2.1 Analytical and Threshold Method

    Analytical method is based on the manual or research experience about data

    sets and set-up our own parameters (thresholds). There is a need for a thorough study of

    falls and ego stage for better understanding and results. What this study focuses on ispresented in this section.

    With fall detection - principles and methods 5.2, many falls end lying on the

    ground, the simplest approach is to detect the lying position, from a horizontal

    inclination sensor. This method is very suitable for monitoring an "isolated a person",

    but less suitable for the detection of falls of an elder person such as the irregular sleeping

    hours. Therefore this method is prone to many "false positives", i.e. detection as falls of

    situations that are not falls.In fall detection system, we can use fall velocity, body orientation, body posture,

    angular acceleration, angular velocity and fall acceleration in the critical phases as the

    determining factors for distinguishing fall and non-falls activities (ADL and ADLS).

    The simplest approach to detect a fall is by detecting the lying position (GPS).

    This is based on the fact that most falls, but eventually not all of them, end up falling in

    this lying positing. Both researches of StanKovic, Noury and Fleury indicate that

    indirect detection of the lying posture during post-fall can be called 5.2, 5.2. That whenthe foot is no longer in contact with the ground also can be another method to detect fall.

    2.2.1.1 Human fall detection algorithm using bi-axial gyroscope sensor

    A threshold-based algorithm to distinguish between Activities of Daily Living

    and falls is described. A threshold-based fall detection algorithm using a bi-axial

    gyroscope sensor is used5.2. With trunk pitch and roll gyroscope are read during

    separate simulatedfall and Activities of Daily Living (ADL) studies, the result will be

    compared through a threshold-based algorithm

    xv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    17/52

    .

    Figure 2-3: Reading trunk pitches and roll gyroscope

    This research focus on studying about the ADL study, Data acquisition set-up,

    sensor location and signal conditioning. In signal conditioning, low-pass filter would be

    used a second-order low-pass Butterworth 2-pass digital filter, with a cut-off frequency

    of 100Hz for each pitch and roll angular velocity signal. The resultant vector was

    derived by from taking the root sum of square of roll and pitch angular velocities.

    Fall detection will apply a threshold to the Peak value of lowest fall and highest

    ADL from the resultant angular velocity signals. The result of recording data will result

    in one of two scenarios:

    Figure 2-4: Overlapping and non-overlapping sampe Fall and ADL data

    In Figure 2-4(A) the peak values recorded ADL will not overlap with the

    recorded fall peak. Falls from ADL which may be distinguished by a

    single threshold, level of which would be placed at the lowest fall peak

    value.

    xvi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    18/52

    In Figure 2-4(B) the peak values recorded ADL will overlap with the

    recorded fall values. In this case applying one threshold is not sufficient to

    distinguish falls from ADL, so continuing to investigate additional aspects

    of the signals is required.

    There are three thresholds to determine for a fall which can be distinguished from

    an ADL. If the resultant angular velocity is greater than 3.1 rads/s (Fall threshold 1), the

    resultant angular acceleration is greater than 0.05 rads/s (Fall threshold 2) and change

    the result in changing trunk angle is greater than 0.59 rads

    (Fall threshold 3), a fall will be detected. The results indicate that falls can be

    distinguished from ADL with 100% accuracy for a total data set of 480 movements.

    Figure 2-5: Bi-axial gyroscope fall-detection algorithm flow chart

    xvii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    19/52

    2.2.1.2 Human fall-detection algorithm using tri-axial

    accelerometer sensor

    Most previous researchers and others combined the devices total acceleration

    from the X, Y and Z axis 5.2. This is called Root Sum of Squares or RSS which ispresented by the formal:

    2 2 2RSS x y z= + + (2.4)

    This is sometimes referred to as the dynamic total acceleration29.81 /dRSS RSS m s=

    Checking the acceleration in early part of the critical phased is the simplest

    algorithm for fall detection. When standing upright, we will have a total acceleration(RSS) of 1G. This value will drop to 0G in free fall until finally reaching impact.

    21 9.81 /g m s=

    This method will use a threshold which is set based on training session. There are

    two thresholds to compare with data from ADLs or other non-fall activities. There are

    called Lower Fall Peak (LFP) and Upper Fall Peak (UFP). In training period, LFP and

    UFP will give the Lower Fall Threshold (LFT) and Upper Fall Threshold (UFT),

    respectively.

    Figure 2-6: Fall detection algorithm operation example for upper and lower fall

    threshold, using an artificial example signal

    Normally, in some case, UFT usually have a better result than LFT.

    xviii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    20/52

    Profiling algorithm in [11] is also to measure the falling edge time and rising

    edge time. With the falling edge time ( FEt ) is the time when the RSS signal last goes

    below the LFT until it reach the UFT. The rising edge time ( REt ) is always a subset and

    smaller than the falling edge time ( FEt ).A specify example, we can imagine that with t0the value of RSS first will go below the Lower Fall Threshold (LFT) and keeps under it

    until 0 200mst+ . The time from 0 200 mst + to 0 500 mst + the RSS value will increase until the

    Upper Fall Threshold (UFT) reached. UFT will reach to peak in 0 500 mst + . 0 5 00mst t+ is

    the galling edge time ( FEt ) and 0 5 0 0 0 2 0 0m st t+ + is the rising edge time. In addition we also

    check for the profile LFT + the falling edge time ( FEt ) and + the rising edge time ( REt

    ).Both LFT and UFT are a prerequisite for using the falling edge time and the risingedge time. However, profiling algorithms only use LFT and UFT can be further

    expanded to also use the falling and rising edge time.

    There are some algorithms that is similar with profiling algorithm such as in

    [12].this algorithm use the time window to find the UFP and LFP and take the difference

    between UFP and LFP ( UFP LFPRSS ). If both of them reach to their respective threshold

    and LFP happens before the UFP. Fall will be flagged.

    UFP LFP UFP LFP RSS t t = with t = time of occurrences (2.5)

    Improving the result of a fall detection system is extreme important so many

    algorithms check the body posture to improve on the specificity. Most of these

    algorithms assume with many case of posture such as making a threshold similar to UFT

    and LFT when the body is sitting or lying or running. We will compute the Lower

    Sitting threshold and the Upper Sitting threshold. After achieving both LFT and UFT 10

    or 15 seconds, if both Lower Sitting threshold and Upper Sitting threshold are achieved,

    fall will be flagged.Another way for falling detection is measuring the velocity which is computed by

    the formula.

    dRSS dt (2.6)

    The velocity will be added in the profiling algorithm and based on the integral

    from the time of the start of a fall until impact.

    xix

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    21/52

    In addition, we can also use the fall index such as in 2.7 or make fall thresholds.

    Especially, we can also make use both in three different types of algorithm.

    First algorithm checks posture and impact. With impact, it will be

    based on Z2, UFT, dUFT or UFP LFPRSS

    .

    The second algorithm uses LFT + UFT (within frame of 1 second)

    or threshold based on Z2 + monitoring.

    The third algorithm uses LFT + threshold with the velocity and

    UFT (within time frame of 1 second) or threshold based on Z2 + monitoring of

    posture. Postured will be detected 2 seconds after the impact using the vertical

    axis acceleration. The sitting and lying posture are usually lower than 0.4G. The

    result of the first algorithm when using Z2 based threshold + posture or UFT isusually 97%. The fall index will be calculated by the formula:

    2

    1

    , , 19

    (( ) ( ) )i

    i k i k i

    k x y z i

    FI A A =

    = (2.7)

    Where x, y, z = acceleration from the X, Y and Z axis

    2 2 2RSS RSS G2

    2G

    dZ

    = (2.8)

    The scalar product of two acceleration vectors can also measured for Posture to

    fin the angle between them. It maybe is between the current gravity vector and the

    reference gravity vector or the gravity and the vertical axis (Ay).

    arccos|| || || ||

    referemcegravity currentgravity

    referencgravity x currentgravity

    = (2.9)

    arccos|| || || ||

    d

    y

    A gravityA x gravity

    = (2.10)

    Some devices have orientation sensors or more accurate gyroscopes which then

    can be used instead 5.2. Both formula 2.5 and 2.11 are used.

    | sin sin cos cos |x z y y z y z

    Orientation A A A = + + (2.11)

    Where x , y , z = the data from the orientation sensor

    xx

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    22/52

    X is the front-back axis, y is the horizontal axis (left and right) and z is the

    vertical axis (up and down).

    2.2.2 Learning Methods for fall detection

    Machine learning can be broadly classified into two fields, supervised learning and

    unsupervised learning. In front, the machine will attempt to identify the groups of

    similar data from a larger dataset. Besides, it also tries to from clusters of data based on

    some criteria such as cost function. It has not prior knowledge of the data layers. It only

    tries to identify natural clusters or groups of data. With supervised learning, it learns

    from the test containing classified data and predicts data layers invisible. With Fall-

    detection system, it will be more natural when using supervised learning techniques.

    Without any analytical algorithm, we can still carry out an intuitive approach

    the development of machine learning based fall detection systems from a training period

    and then classification. However, It may be is necessary to establish criteria for

    classification.

    The most important in supervised learning is the quality of accurate and

    exhaustive of training data. Because of these reasons, in the period of training, it is

    important to be able to simulate how falls as close and how the real falls for other groups

    of users who will use the fall detection system.

    With some machine-learning algorithm out there, it may be cause confusing

    about selecting the right one for certain applications. Even thought, some research also

    studies about comparing the result of them to try to choose the best algorithm 5.2,5.2 To

    sum up, a developer can be chosen an application method that is suitable with his (her)

    own discretion. An evident with fall detection system is a sensitive application. It is

    advised to compare and test between a numbers of algorithms to try picking the best

    choice.Ralhan in University of Saskachewan compared five different supervised

    learning methods for fall detection system 5.2. Naive Bayesian classifier, Radial basis

    Function (RBF), C4.5 and Ripple down Rule Learner, Support Vector machine are used.

    Eight scenarios (four falls, three ADLS and one near fall) are used to getting data from

    test the algorithms against. With Nave Bayes classifier, the highest result is 97% and

    taking the least time for building model and then it was chosen as the best choice for

    research.

    xxi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    23/52

    In addition, Fall detection by embedding an accelerometer a method that of Ralhan

    in University of Saskachewan used with machine learning algorithm 5.2. They used the

    group integrated a device with tri-axial accelerometer, a MCU device and some other

    peripherals with a mobile phone. One class SVM for preprocessing the signals and KFD

    (Kernel Fisher Discriminant) and K-NN (Nearest neighbor) was used with fall detection

    for precise classification. The result of research is 92.3% for a limited number of cases.

    xxii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    24/52

    Chapter 3

    As mentioned in Chapter 2, we have presented some fall detection system that

    uses various approaches in the study. Their disadvantages and advantages were

    discussed. There is need for a precision of fall detection system that wishes train andtest. Because of this, studying for falls is an important issue with the research.

    With intent and purposes of the system that we want to make. We define a fall to

    be a sudden change of body position coming to rest on the ground, it does not include

    intentional change, and then is inactivity. This means that the most serious falls where

    user loses a balance after hitting the ground. They also lose the ability for getting help.

    In particular, the system checks a sudden change in acceleration towards a

    negative value. This is a sign of a change in a normal operation. This is similar asentering the critical phase from pre-fall. After that a change of acceleration toward a

    positive value will be checked in the system. This value becomes higher with the harder

    impact. This means that the kind of surface of impacts have occurred on many issues.

    How quickly the user is falling down also change this value.

    Since we have determined that our research only recognize serious falls, we can

    simplify our definition of a fall and non-fall or the acceleration towards the ground after

    a hard impact and the time of inactivity.

    xxiii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    25/52

    As we only test the common types of falls by using minimal. We will not assume

    any particular group of users. It will be wrong to claim that our research is made

    specifically for a group such as the elderly. Elderly people fall in so much more ways

    than we define. They can be fall in anywhere such as walking, lying or from bed, etc.

    For example Boyle and Krunanithi (2008) defined specifically nineteen different ways

    of falls observed from the study about elderly people 5.2.

    3.1 Study of fall

    Definition a fall

    Many definitions about a fall are given. This is because fall for one specific

    group of users such as young people, elder people or different other groups. The thesisfocuses on distinguishing fall and non-fall activities (it may be called Activity of Daily

    Livings or ADLs). Now as mentioned, types of fall of each group of people always are

    different (elderly people or childetc).

    Figure 3-7: The stage of fall

    In our research, we will distinguish between two kinds of falls. The first place is

    falls which are caused by external factor (such as behavioral or environmental). The

    second place is falls that caused by internal factors (such as biology).

    Lying down/ Standing/ Walking

    Fall Forward/Backward/Left/Right

    xxiv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    26/52

    o For accidental falls

    o For non-accidental falls

    Near FallCause of falling

    The normal changes of aging, like poor vision and poor hearing. They can make

    you more likely to fall.Diseases and physical conditions can affect your strength and

    balance.

    There are some risk factors for cause of falling. In this thesis, we divide these

    factors into intrinsic (biology) and extrinsic (behavioral and environmental). We use

    both of factors to detect falling.

    Firstly, biological factors may be are:

    1. Age

    2. Medical condition such as Parkinsons disease, patients.

    3. Muscle Weakness

    4. Visual impairment.

    5. Foot problem.

    Secondly, Behavioral and Environment factors are:

    1. Sedentary

    2. Medication intake.

    3. Alcohol misuse.

    Beside two types of groups for falling, we also research some trends of falls.They are as follows:

    1. Time

    2. Climate/weather

    3. Location

    4. Race

    5. Depression

    xxv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    27/52

    Phases of falling

    There are four phases of falling: the pre-fall, the critical phases, the post-fall phase

    and the recovery phase

    Figure 3-8: The four phases of a fall event

    The pre-fall phase corresponds movement of daily life, sometimes with

    sudden movements directed towards the ground such as sitting down or

    bending over. These activities should not create an alarm with a fall

    detection system.

    The critical phase, corresponding to the fall, is highly short. This phase

    can be detected by the motion of the body toward the ground or by the

    shock of impact with the floor

    The post-fall phase is usually characterized by a real person on the

    ground immediately after the fall. It can be detected by a lying position or

    by an absence of significative motion

    A recovery phase may eventually occur if the person can stand alone or

    with the help of another person

    3.2 The simulated-fall study

    The simulated fall study related to healthy young subjects and carrying out under

    the supervision. Tri-axial accelerometers signals recorded from the trunk and thigh in a

    fall event simulation. Each subject performed eight different kinds of fall and each type

    was repeated three times.

    The fall types used in the testing process for the current study were selected to

    simulated common types of fall in elderly people. The most common causes of falls are

    the trips, slips, and loss of balance. ONeill et al.5.2indicated that 60% of falls in older

    xxvi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    28/52

    people were falling in the forward direction. Laterally directed fall also were a major

    pose of threat. Laterally directed fall causes the bad impact such as the potential to

    fracture every time it happens. Therefore, falls from all directions should be considered

    when validating fall detection system. We also should attempt to make mimicking the

    realistic falls. Thus, simulation of performance were backward falls, lateral falls right,

    lateral fall left and forward falls.

    3.3 The ADL study (Activity of Daily Living)

    The second study involved older people performing ADL, in their own homes,

    while equipped with the same sensor configuration. Ten subjects of the elderly

    community, four females and six males, were monitored. They ranged in age from 60 to80 years old. Each ADL was performed three times by each older person. The ADL

    were a task that could produce impacts or sudden changes in movement and result of a

    mistake caused by a threshold-based fall detection algorithm

    The second study involved every activity of living from children to elderly

    people. Each ADL was performed three times by each person. To evaluate fall detection

    algorithm, Bourke, OBrien and Lyons showed some types of ADL in their research 5.2:

    1 Sitting down and standing up from an arm-chair

    2 Sitting down and standing up from a kitchen-chair.

    3 Sitting down and standing up from a toilet seat.

    4 Sitting down and standing up from a low stool

    5 Getting in and out of a car seat

    6 Sitting down on and standing up from a bed

    7 Lying down and standing up from a bed

    8 Walking 5m

    9 Cycling

    xxvii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    29/52

    Chapter 4

    In the period of studying for fall detection and the installation of experiment, we

    found that the fall detection algorithms based on threshold still have the following

    limitations:

    They have not been built a specific training data on typical actions falls.

    The techniques of experimental process based on the characteristic time

    domain and or frequency domain were not mentioned. They have not been

    proper interest in these methods.

    In addition as mentioned in the previous chapter about the theory of fall

    detection. In the data collection, human undesired movement can cause

    large changes in intensity dramatically at that moment. It affects directly

    on the result obtained when comparing with the threshold.

    To solve the above disadvantages, we proposed the method to extract feature

    combined building layer model for fall detection system. In particular, the model will be

    proposed for using extracted techniques based on the information on the time domain

    and the characteristic value in accelerometer sensor. The following thesis will present

    general model for classification

    xxviii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    30/52

    Figure 4-9: The general model cycle

    The components constitute the layer model using in this thesis. This model is

    based on the criteria that the thesis proposed to decrease limitations of the above

    methods. Accelerometers data will be collected directly from the user through specific

    cases which described. Next part is to expand intensity and assign label for datasets. In

    this part, some other information of accelerometer data on the time domain will be

    extracted which are used in the classifying phase.

    4.1 Accelerometer Data collection

    In this section, we introduce first t how the data collected. After that, we will

    present fall detection performance achieved by different algorithms. We collected data

    of falls from different directions (x, y, z axes) and different environment (Bedroom,

    kitchen, outdoor garden, living room) with other cases of fall. The data from activities of

    daily living (ADL) such as walking, jogging, sitting or standing are also be collected.

    This below sample data acquired from accelerometer and fall annotation for a single fall:

    xxix

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    31/52

    Table 4.1: sample data acquired from acceleromter

    x-axis y-axis z-axis time

    0.459687 1.37906 9.80665 572235910984060.459687 1.532289 9.346964 57223619296739

    0.153229 1.225831 9.95987 572236779434050.459687 1.225831 9.040505 57223741578404-0.30646 0.459687 10.72602 572237994167390.153229 0.459687 9.500193 572238625934080.153229 0.153229 9.500193 572239211800750.153229 -0.45969 10.5728 572239823300750.153229 -0.30646 9.500193 57224039275074-0.15323 -0.76614 9.193734 57224100655074-0.1533 -1.53229 9.346964 57224161960072

    0.612916 1.225831 9.959879 57223529871740

    0.459687 1.37906 9.8066 572235910984060.459687 1.532289 9.346964 572236192967390.153229 1.225831 9.959879 57223677943405

    Figure 4-10: fall annotation for a single fall

    The experiments with groups of real persons are conducted. Both activities of

    daily living and falls are tested and training. However, we cannot test falls with elderly

    people. We got data from 10 participants who are student from 18 to 35 years old ( body

    mass from 50 to 65kg and height from 1.62 to 1.76m) and 5 children from 7 to 12 years

    old.(body mass from 25 to 40kg and height from 1.4 to 1.55 m) .

    In getting data for fall detection system, the cell phone was put in the shirt pocket

    of participants. In each case of position, every participant falls 7 times in different

    directions and environment. To sum up, we recorded 5 examples of the behavior from

    xxx

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    32/52

    15 persons and collecting Activities of daily living for 2 or 3 minutes from each person

    with some cases such as sitting, running, lying or walking. Mitja and Igone 5.2 also gave

    some multiple activities:

    3 x 15 recordings of falling, consisting of standing/ walking, falling andlying. Fall detection is one of the main goals of our project.

    3 x 10 recordings of lying down, consisting of standing/walking, lying

    down and lying. Lying down is same as falling down, so we wanted to

    verify whether the classification can distinguish between the two.

    3 x 10 recordings of sitting down, consisting of walking, sitting down and

    sitting. Sitting down may also resemble falling and is a common action .It

    is important for the analysis of the users behavior.

    3 x10 recordings of walking. Walking is also popular and we wanted

    clearly examples of it to test the classification. Recognizing word using

    trial cutting and beam search

    4.2 Algorithm for Fall Detection

    This subsection describes a framework for processing data. It is namely pre-noise

    processing data, attribute computation and feature selection. Especially, Feature

    selection is an important prior step to any classification problem which reduces the

    dimensionality and thus the amount of data required for training. They will be presented

    as follows:

    4.2.1 Pre-noise processing data

    Because the data is received from sensor (usually is noisy data), so the techniquefor filtering data was applied. There are several techniques that perform signal filter. In

    this research, we applied a low-pass and high-pass filter.

    A simple low-pass filter for the time domain is a smoothing function. In other

    work, the signal is smoother and less dependent on short changes. We use low-pass filter

    to reduce the influence of sudden change on the accelerometer data.

    xxxi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    33/52

    It also can filter a series such as the low-frequency variations and reduced and

    the high-frequency is unaffected. This type is called a high-pass filter. This is

    particularly important in the acceleration data. This type also allows us to remove the

    gravity component and take into consideration the sudden change in acceleration.

    The below algorithm shows the low-pass and high-pass filter that we use in this

    research. It uses a low-value filtering factor to generate a value that uses 20% of

    the unfiltered acceleration data and 80% of the previously filtered value. This factor

    was chosen empirically.

    Figure 4-11: Low-pas and high-pass filter algorithm

    4.2.2 Attribute computation

    In this section, we describe the process of computing attributes. After that, these

    attributes are combined to create the final attribute vector for using in the machine

    learning classification. Most attributes were computed by using the technique ofoverlapping sliding windows. Sliding window is a common approach to solve the

    problems of activity recognition. Normally, the algorithms do not attempt to recognize

    the pattern that is received from sensor but trying to recognize some patterns in the data

    which is over time in the window.

    When analyzing series of time, we chose a window size of 10 which is one

    second time interval. We decided for one-second time because some transitional

    activities usually last from one to five seconds.

    xxxii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    34/52

    4.2.3 Feature Extraction

    To detect fall based on accelerometer sensor with machine learning algorithm, we

    will use a sliding window to transform stream of acceleration data into instances for

    machine learning. The following attributes were derived from the data within slidingwindow. We will against on some method to calculation:

    Length of the acceleration vector

    The variance of the length of the acceleration vector within the window

    The average acceleration along the x, y and z axes within the window

    The speed of change in acceleration between the maximum and minimum

    along the x, y, z axes. The maximum and the minimum acceleration along the x, y, z-axes

    The acceleration vector changes (AVC)

    The accelerometer inclination angles

    4.2.3.1 Length of the acceleration vector

    The first computed attribute is length of the acceleration vector. It is a simple butit is very useful attribute. Moreover, it is also used further in the process of extraction

    the new attributes. It is not used as a separate attribute in the final vector because of the

    sliding window technique. Its definition is:

    2 2 2RSS x y z= + +

    xxxiii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    35/52

    Figure 4-12: RSS plot of a forward fall

    During static posture this attribute is constant with the value equal to the Earchs

    gravity (RSS = 1g). In dynamic activities the acceleration vector is changing the

    direction and its length.

    4.2.3.2 The variance of the length of the acceleration vector within the window

    With these characteristics, based on the average, maximum and minimum

    acceleration along the y-axis, we compute the variance of the length of the acceleration

    vector. Especially with z-axis, the speed of change was also calculated.

    With the variance with in a window, it will be defined as follows:

    2

    2 1

    ( )n

    i

    i

    a a

    N

    =

    = (4.1)

    With N is the number of acceleration data within the window, ia is the length of

    the thi acceleration vector and a is the average length of the acceleration vector of the

    person)4.2.3.3 The speed of change in acceleration along the z-axis

    The speed of change in acceleration along the z-axis within a window was

    defined as follows:

    max( ) min( )

    tan(max( ) (min( ))

    z zz

    z z

    a aspd

    t a t a

    =

    (4.2)

    For this attribute, the value of raw materials for the length of the acceleration

    vector is used instead of low-pass value .Max ( za ) and min ( za ) are the maximum and

    xxxiv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    36/52

    minimum acceleration along the z axis within the window, and t (max ( za )) and t (min (

    za )) are the time stamp of the data.

    4.2.3.4 The acceleration vector changes (AVC)

    When the person's body is static, single accelerometer response only to the

    gravity, producing a constant length of acceleration is 1g. In the moving accelerations

    produce a changing in acceleration signal and drastically change the motion. Using

    changes in the acceleration vector, an attribute is calculated to detect the motion

    accelerometer: Acceleration Vector Changes (AVC). The AVC value of this attribute

    increased as the acceleration (walking, going down, stand up, etc ...). This attribute take

    into consideration the data from the current window (ten data samples). It combines ten

    different length of the vector of length acceleration vector and divided the sum by thetime interval (one second) of data. The AVC is computed as followed:

    1

    1

    0

    | |n

    i i

    i

    n

    length length

    AVCT T

    =

    =

    (4.3)

    0T is the time stamp for the first data sample in the window, and nT is the time

    stamp of the last data sample. The fall can be detected by using this attribute. With this

    attribute, the movement of people can be detected: it distinguishes static from thedynamic active. For this attribute, the value of raw materials for the length of the

    acceleration vector is used instead of low-pass value. The reason for this is that we are

    more interested in small changes in the acceleration signal and the smooth of low-pass

    filter.

    4.2.3.5 The accelerometer inclination angles

    Other important features to be recognized a static body posture are the

    orientation angles of the accelerometer. The orientation angles are calculated as the

    angles between the actual acceleration and each of the axes (x, y and z axes). (Figure 4-

    13)

    xxxv

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    37/52

    Figure 4-13: Accelerometer inclination angle

    For instance, the angle x between the acceleration vector and the x axis(perpendicular to the ground) is computed as follows:

    2 2 2

    arccos( )x

    x y z

    ax

    a a a

    =

    + + (4.4)

    Where the value ax, ay, az represent the actual acceleration vector. In this

    attributes, lower-pass filter should be used. Because of the changes of the angle is less

    and less variation. Without the low-pass filter, the angles were sensitive angles to each

    small change of the accelerometer. These angles improve the classification of activities

    that have different accelerometer angle inclinations.

    4.3 Fall detector classification

    After filtering the data, we use a sliding window to transform the stream of

    acceleration data into instances for machine learning. The attribute of acceleration vector

    will be computed. Then we will be divided into the final vector for specific actions

    which tries to classify the appropriate fall and non-fall of the user. This process is

    indicated in Figure 4-14. The classification will be performed with SVM function on the

    software toolkit Weka.

    Figure 4-14: Fall Detection Flow Chart

    xxxvi

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    38/52

    Chapter 5

    This chapter provides some measures, namely sensitivity, specificity, F-score for

    evaluating performance of the classification methods. We supply two evaluation criteria

    to measure our system presented in Chapter 4.

    5.1 Data Collection

    In this experiment, the subject was divided into 2 groups:

    Ten young subjects (6 male and 4 female, age 256 years, body mass 5311.5 Kg

    and height 16610.5 cm)

    Ten elderly subjects (6 male and 4 female, age 646 years, body mass 525kg, and

    height 1645cm).

    To compare the result of fall detection by variety of subject groups to perform falls and non-fall

    (ADL), the experiments were divided into 2 types:

    Type A: Young subject perform both ADL and simulated falls

    Type B: The young subjects performed simulated falls and the elderly people

    performed ADL.

    xxxvii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    39/52

    5.2 Test Scenario

    To have database to train and test the character recognizer, we designed a number

    of this scenario was designed to investigate the events which can be difficult to

    recognize such as fall and non-fall. It includes 9 different events which are showed inTable 5.2. They are recorded in single recordings that it interspersed with short periods

    of walk, each record lasted from 3 to 5 minutes. An example for video of the event can

    be view in http://dis.ijs.si/confidence/iaai.html

    Table 5.2: Events Sequence in the test Scenario

    No Description Fall Ending position1 Sitting down normally on the chair No

    2 Tripping, landing normally on the

    bed

    Yes

    3 Lying down normally on the bed No

    4 Falling slowly( trying to hold onto

    furniture), landing flat on the ground

    Yes

    5 Sitting down quickly on the chair No

    6 Falling when trying to stand up,landing sitting of the ground

    Yes

    7 Lying down quickly on the bed No

    8 Falling slowly when trying to stand

    up(trying to hold onto furniture),

    landing sitting of the ground

    Yes

    9 Searching for something on the

    ground on all fours any lying

    No

    Some types of falls were selected from a list of typical falls. As shown in the

    choosing on related work. Accelerometer sensor can accurately detect typical falls.

    Because of this, we gave a fall (such as event 2) to prove that the system can recognize

    .In addition, we also choose three atypical falls (event 4, 6 and 8) to test the use of

    information in each circumstance. A specific example that a person is not expected to lie

    or sit on the ground (in contrast with the bed or the chair). They are atypical in speed

    (event 4 and 8) and starting/ending posture (event 6 and 8). Each scenario was repeated

    xxxviii

    http://dis.ijs.si/confidence/iaai.htmlhttp://dis.ijs.si/confidence/iaai.html
  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    40/52

    3 times with each subject. Therefore, there were 300 sequences of data (120 fall data and

    180 ADL data)

    5.3 Algorithm sensitivity

    After all acceleration signals in three axes were acquired, they were converted tothe resultant acceleration by the average of length vector, the root sum of square, the

    covariance and the speed of change. These values are also based on the maximum and

    the minimum peak resultant acceleration was used for classifying falls and non-fall

    (ADL) in two types of data i.e. type A and B from the experiment.

    We designed five sets of attributes describing the behaviors of the subject. The

    reference attributes are presented in the reference coordinate system, which is fixed with

    environment of users. The attributes of the subjects body are also shown in a coordinatesystem.

    5.4 Performance measures

    Fall detection is either positive if the detector properly recognizes a fall event or

    negative if it does not. The quality of fall detection cannot be evaluated from a single

    test; instead, it is necessary carry out from a series of test. It will include four possible.

    True positive (TP): fall occurs, the device detect it

    True Negative (TN): a normal movement (non-fall) is performed. The device

    does not declare a fall.

    Fall Positive (TP): The device declares a fall, but fall does not occur

    Fall Negative (FN): a normal movement (non-fall) is performed. The device

    declares a fall.

    Based on four situations, we proposed two criteria to evaluate the fall detection

    system:

    Sensitivity is the capability to detect a fall. It can be expressed:

    ensTP+FN

    TPS itivity =

    Specificity is the capability to detect an ADL (Activity of Daily Living). It

    can be expressed:

    xxxix

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    41/52

    TN

    SpecificityTN FP

    =

    +

    5.5 Experimental results of the fall detection system

    In this thesis, based on experience, after collecting data and feature extraction, wepreformed the experiment with other data set by LibSvm function on the software toolkit

    Weka.. Below this is the result of experiments with young subjects performed both ADL

    and simulated falls.

    Table 5.3: Effect of body profile on the performance

    Volunteers Age(years)

    Weight(kg)

    Height(cm)

    Sensitivity% specificity% F-Measure%

    Subject 1 22 53 169 81 84.3 87.1

    Subject 2 19 54 168 83.6 87.2 84.7

    Subject 3 33 58 164 83.6 84 83.5

    Subject 4 25 56 165 93.1 93.2 93.1

    Subject 5 28 63 174 85.5 85.2 84.8

    Subject 6 21 55 170 87 86.4 86

    Subject 7 25 59 166 93.8 93.8 93.8

    Subject 8 23 60 171 89 89.4 89.4

    Table 5.3 shows the system achieves a respected F-measure. The result is not

    high because the created characteristics are not large enough to cover all patterns for

    classification.

    With the ratio 80/20 divided at random in the datasets which gives us a high

    result. Although it is not stable because it is dependent on the accuracy of collected data.

    Following this, in our experiments with the data obtained, after analyzing the datasets.

    We divided to train and test data into separate files with the specific percentage. They

    are used to train the one-class SVM model, and other to test the model, as shown in

    xl

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    42/52

    Table 5.4. The result is showed in Table 5-4. The results show that this model can detect

    the fall effectively.

    Table 5.4: The samples for each category

    Category total samples samples fortraining

    sample fortest

    demonstrated by

    1 210 150 60 140 younger, 70 elder

    2 130 90 40 younger

    3 90 70 20 50 younger , 40 elder

    4 150 100 50 120 younger , 30 elder

    5 100 70 30 younger

    6 50 25 25 40 younger, 10 elder

    Table 5.5: The test result

    Category 1 2 3 4 5 6

    Accuracy (%) 82.64 93.3 87.59 88.7 95.4 92.6

    When using an accelerometer for fall detection, the machine learning based

    methods somewhat outperformed threshold algorithm. Considering the simplicity of

    these methods, threshold may be simpler than machine learning methods, but the related

    work shows that in some case, machine learningbased methods gave the better result

    because this may seem surprising, but the related work shows that threshold-based

    methods were apparently able to learn the pattern of acceleration during the falling.

    xli

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    43/52

    Chapter 6

    Conclusion and Future

    work

    This thesis studies built fall detection system based on accelerometer sensor. We

    studied the fundamental knowledge about fall detection such as data feature extraction,

    machine learning models, support vector machine and how to combine them.

    This thesis has achieved the following results: Study to build a model of fall detection based on accelerometer sensor

    Studying and learning the typical case of falls, the processing techniques and the

    characteristic of the accelerometer sensor are used.

    About theory and methods, in terms of the thesis, we explore several detection

    techniques identify different falls including pre-processing techniques

    accelerometer sensor, extract selected characteristic, standardized features and

    classification techniques using in signal classification problems.

    Experimentally, based on the study of acceleration sensor and the method of

    digital signal processing, we have built and installed four experimental models.

    In particular, we proposed a model to improve the accuracy of fall detection

    based on accelerometer sensor to ensure objectivity, the proposed model was

    compared with the evaluation results of other fall detection algorithms in

    previous studies.

    xlii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    44/52

    However, this thesis still exist some limitations:

    The size of the test data is not large enough for a fully evaluated and general

    methods and experimental models.

    Accelerometer sensor with different users is different Therefore, choosing thecharacteristics vector and the width of the window side corresponding to the

    intersection point of the user are not be automated.

    In the future, we will continue to improve the recognition quality and try to fix the

    limitations through the development of a standard signal from many different users more

    fully

    xliii

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    45/52

    Fall Scenarios

    Ngo Duy Kien

    Bibliography

    [1] Li,Q.,Stankovic,J.,Hanson,M.,Barth,A.,Lach,J.,&Zhou,G.2009.Accurate,fast fall

    Detection using gyroscopes and accelerometer-derived posture information. In

    Wearable and Implantable Body Sensor Networks, 2009. BSN2009. Sixth

    International Workshop on, 138143

    [2] Noury,N.,Fleury,A.,Rumeau,P.,Bourke,A.,Laighin,G.,Rialle,V.,&Lundy,J.2007.Fall

    detection-principles and methods. In Engineering in Medicine And

    BiologySociety,2007. EMBS2007.29th Annual International Conference of the

    IEEE,16631666.

    [3] Mitja, L. &Bostjan, K. Fall detection and recognition with machine learning.

    Technical report, Joef Stefan Institute, Department of Intelligent Systems, 2009.

    http://dis.ijs.si/mitjal/documents/Fall_detection_and_activity_recognition_with_mac

    hine_learning-Informatica-09.pdf.

    [4] Noury, N., Fleury, A., Rumeau, P., Bourke, A., Laighin, G., Rialle, V., & Lundy, J.

    2007. Fall detection - principles and methods. In Engineering in Medicine and

    Biology Society, 2007. EMBS 2007. 29th Annual International Conference of the

    IEEE, 1663 1666.

    [5] Zhang, T., Wang, J., Liu, P., &Hou, J. 2006. Fall detection by embedding an

    accelerometer in cell phone and using kfd algorithm. In International Journal of

    Computer Science and Network Security, Vol. 6 no. 10 pp 227284.

    [6] Ralhan, A. S. A study on machine learning algorithms for fall detection and

    movement classification.Masters thesis, University of Saskatchewan,

    http://library2.usask.ca/ theses/available/etd-12222009-144628/, 2009.

    [7] A.K.Bourke,J.V.OBrien,G.M.Lyons Evalutaion of a threshold-based tri-axial

    accelerometer fall detection algorithm

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    46/52

    Fall Scenarios

    Ngo Duy Kien

    http://www.eee.nuigalway.ie/documents/go_gait_posture_2007.pdf

    [8] Ubisense. http://www.ubisense.net, 2008-09-15

    [9] Maybeck, P. S. Stochastic models, estimation, and control. Mathematics in Science

    and Engineering 141, 1979.

    [10] Dai,J.,Bai,X.,Yang,Z.,Shen,Z.,&Xuan,D.292010.

    Perfall: A pervasive fall Detection system using mobile phones. In Pervasive

    Computing and Communications Workshops (PERCOM Workshops), 2010 8th IEEE

    International Conference on.

    [11] Bourke,A.,vandeVen,P.,Gamble,M.,OConnor,R.,Murphy,K.,Bogan,E.,McQuade,E.,

    Finucane, P., Laighin, G.,&Nelson,J.2010.Evaluationofwaist-mountedtri-axial

    Accelerometer based fall-detection algorithms during scripted and continuous

    unscripted activities. Journal of Biomechanics, 43(15), 30513057

    [12] Jantaraprim,P.,Phukpattaranont,P.,Limsakul,C.,&Wongkittisuksa,B.may2010.

    Improving the accuracy of a fall detection algorithm using free fall

    characteristics. In Electri-Cal Engineering/Electronics Computer

    Telecommunications and Information Technology (ECTI-CON), , 501504.[13] A threshold-based fall-detection algorithm using a bi-axial gyroscope sensor

    http://www.eee.nuigalway.ie/documents/go_med_eng_phys_2008_a.pdf

    [14] Fall detection Using 3D Head Trajectory Extracted From a Signal Camera Video

    Sequence

    http://www.computer-vision.org/4security/pdf/montreal-fall_detection.pdf

    [15] Boyle, J. &Karunanithi,M.2008. Simulated fall detection via accelerometers. In

    Engineering in Medicine and Biology Society, 2008.EMBS 2008.30th AnnualInternationalConference of the IEEE, 12741277.

    [16] Dai,J. ,Bai,X., Yang,Z., Shen,Z., &Xuan, D.292010.Perfalld: Apervasive fall

    detection system using mobile phones. In Pervasive Computing and Communications

    Workshops (PERCOM Workshops), 2010 8th IEEE International Conference on.

    [17] Mitja LUTREK1, Matja GAMS1, Igone VLEZ2, IST-Africa 2009 Conference

    Proceedings Paul Cunningham and Miriam Cunningham (Eds) IIMC International

    http://www.eee.nuigalway.ie/documents/go_med_eng_phys_2008_a.pdfhttp://www.computer-vision.org/4security/pdf/montreal-fall_detection.pdfhttp://www.eee.nuigalway.ie/documents/go_med_eng_phys_2008_a.pdfhttp://www.computer-vision.org/4security/pdf/montreal-fall_detection.pdf
  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    47/52

    Fall Scenarios

    Ngo Duy Kien

    Information Management Corporation, 2009. Posture and Movement Monitoring for

    Ambient Assisted Living

    [18] Caruana,R.&Niculescu-mizil,A.2006.An empirical comparison of supervised

    learning algorithms. In InProc.23rdIntl.Conf.Machine learning(ICML06),161

    168.

    [19] Caruana,R.,Karampatziakis,N.,&Yessenalina,A.2008.Anempirical evaluation

    of supervised learning in high dimensions. In International Conference on

    Machine Learning (ICML) 96103.

    [20] ONeill TW, Varlow J, Silman AJ, Reeve J, Reid DM,Todd C,etal.

    Age and sex in fluences on fall characteristics. Ann Rheum Dis1994;53(11):7735

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    48/52

    Fall Scenarios

    Ngo Duy Kien

    Appendix A

    A. Fall Scenarios

    1. Backward Fall

    2. Fall to the Left

    3. Fall to the right

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    49/52

    Fall Scenarios

    Ngo Duy Kien

    4. Fall to the right

    5. Forward fall collapse

    6. Forward fall collapse

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    50/52

    B, ADL (Non-Fall)

    1. From standing position, sit down (normal speed) on chair , remain 5 seconds and

    then stand up

    2. Kneel down and pick up an item

    3. Standing position, bend-down and pick up an item on the floor. Rise-up

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    51/52

    4. Standing position, quickly fall down facing the floor, push up 3-5 times then rise up

    5. Lie down and remain 5 seconds. Rotate 180 degree, remain 5 second , then rise up

    6. Standing position, back to original position

    7. Standing position and jump

    8. Standing position and run from A to B.

  • 7/31/2019 Ngo Duy Kien _Thesis Version 10_25

    52/52

    9. Lie down on the table

    10. Climb up on the table, remain 5 second and then jump down