Fall Detection based on Movement and Smart Phone Technology Vo Quang Viet ECE, Chonnam National University Gwangju, South Korea Email: vietquangvo@gmail.com Gueesang Lee ECE, Chonnam National University Gwangju, South Korea Email: gslee@chonnam.ac.kr Deokjai Choi ECE, Chonnam National University Gwangju, South Korea Email: dchoi@jnu.ac.kr Abstract—Nowadays, recognizing human activities is an important subject; it is exploited widely and applied to many fields in real-life, especially health care or context aware application. Research achievements are mainly focused on activities of daily living which are useful for suggesting advises to health care applications. Falling event is one of the biggest risks to the health and well being of the elderly especially in independent living because falling accidents may be caused from heart attack. Recognizing this activity still remains in difficult research area. Many systems which equip wearable sensors have been proposed but they are not useful if users forget to wear the clothes or lack ability to adapt themselves to mobile systems without specific wearable sensors. In this paper, we develop novel method based on analyzing the change of acceleration, orientation when the fall occurs. In this study, we recruit five volunteers in our experiment including various fall categories. The results are effective for recognizing fall activity. Our system is implemented on Google Android smart phone which already plugged accelerometer and orientation sensors. The popular phone is used to get data from accelerometer and results show the feasibility of our method and contribute significantly to fall detection in Health care. Keywords— Fall Detection; Activity Recognition; Activity of Daily Living (ADL); Smart phone I. INTRODUCTION Activity recognition is researched to determine the action or states of the user by analyzing sensor data. With simple human activities such as walking, running, a large number of classification methods have been investigated. Users use wearable accelerometer or bring mobile phone equipped with accelerometer. In this way by exploiting X, Y and Z value of accelerometer we can infer physical activities of users. Many studies are based on this method to contribute classifiers. Specially, the fall is a very risky factor in the elderly people’s daily living, especially the independent living, since it often causes serious physical injury such as bleeding, and centre nervous system damages. A 1/3 rate of the persons aged over 65 has been reported to occur at least once per year [1]. If the emergency treatments were not in time, these injuries could result in disability, paralysis, even death. The current fall detection methods can be basically classified into three types based on data [1]: -Video data: The video based system captures the images of human movement, then determines whether there is a fall event or not based on the variations of some image features and using machine learning algorithms [2]. -Acoustic data: Detecting a fall via audio signals analysis. -Wearable sensor data: Embedding some micro sensors into clothes, or girdle, shoes, plug on foot, etc. to monitor the human activities in real-time, and find the occurrence of a fall based on the changes of some movement parameters [3]. However, these approaches have some weak points as narrow scope of video system, poor accuracy of audio system, and inconvenience of wearable sensors system. Recent smart phones are well equipped with useful sensors including accelerometers and orientation. With maturation of Internet and rapid development in mobile communication, device could bring enhanced services to the person especially Health Care center. Therefore, we focus on developing a fall detector based on mobile handset. Recent reports on posture tell that people have popular habit using mobile phone by putting the phone next to their ears for listening or calling, holding it by hand for typing short message service (SMS) or playing game, and putting mobile phone in bags, pants pocket or check pocket [4]. A fall event can happen in those cases. Therefore, we separate them into categories to easily observe the change of acceleration and orientation as in Table 1. TABLE I. THE POSTURES OF USING PHONE Mobile position Description Category 1 Hold by hand while typing SMS Category 2 Hold by hand while listening Category 3 In chest pocket Category 4 In pants pocket Accelerometer has been proposed being suitable for falls detection, but there still remains basic restrictions. The basic approach was published in [5]. In that approach, a change in orientation that occurs immediately when user have bumped into something while walking is indicative of a fall event. The common and simple methodology for fall detection is using a tri-axial accelerometer with threshold algorithms [6]. Such algorithms simply raise the alarm when 978-1-4673-0309-5/12/$31.00 ©2012 IEEE 163