• 센서학회지
• /
• 제25권2호
• /
• pp.79-85
• /
• 2016

This paper proposes a threshold-based fall recognition algorithm to discriminate between falls and activities of daily living (ADL) using a tri-axial accelerometer and a bi-axial gyroscope sensor mounted on the upper sternum. The experiment was executed ten times according to the proposed experimental protocol. The output signals of the tri-axial accelerometer and the bi-axial gyroscope were measured during eight falls and eleven ADL action sequences. The threshold values of the signal vector magnitude (SVM_Acc), angular velocity (${\omega}_{res}$), and angular variation (${\theta}_{res}$) parameter were calculated using MATLAB. From the preliminary study, three thresholds (TH1, TH2, and TH3) were set so that the falls could be distinguished from ADL. When the parameter SVM_Acc is greater than 2.5 g (TH1), ${\omega}_{res}$ is greater than 1.75 rad/s (TH2), and ${\theta}_{res}$ is greater than 0.385 rad (TH3), these action sequences are recognized as falls. If at least one or more of these conditions is not satisfied, the sequence is classified as ADL.

• The Journal of Korean Physical Therapy
• /
• 제22권6호
• /
• pp.59-63
• /
• 2010

Purpose: Driving a car is an essential component of daily life. For safe driving, each driver must perceive sensory information and respond rapidly and accurately. Brake response time (BRT) is a particularly important factor in the total stopping distance of a vehicle, and therefore is an important factor in traffic accident prevention research. The purpose of the current study was (1) to compare accelerometer. BRTs analyzed by three different methods and (2) to investigate possible correlations between accelerometer-BRTs and foot switch-BRTs, which are measured method using a foot switch. Methods: Eighteen healthy subjects participated in this study. BRT was measured with either a tri-axial accelerometer or a footswitch. BRT with a tri-axial accelerometer was analyzed using three methods: maximum acceleration time, geometrical center, and center of maximum and minimum acceleration values. Results: Both foot switch-BRTs and accelerometer-BRTs were delayed. ANOVA for accelerometer BRTs yielded significant main effects for axis and analysis, while the interaction effect between axis and analysis was not significant. Calculating the Pearson correlation between accelerometer-BRT and foot switch-BRT, we found that maximum acceleration time and center of maximum and minimum acceleration values were significantly correlated with foot switch-BRT (p<0.05). The X axis of the geometrical center was significantly correlated with foot switch-BRTs (p<0.05), but Y and Z axes were not (p>0.05). Conclusion: These findings suggest that the maximum acceleration time and the center of maximum and minimum acceleration value are significantly correlated with foot switch-BRTs.

• 한국전문물리치료학회지
• /
• 제23권1호
• /
• pp.87-93
• /
• 2016

Background: Posture balance control is the ability to maintain the body's center of gravity in the minimal postural sway state on a supportive surface. This ability is obtained through a complicated process of sensing the movements of the human body through sensory organs and then integrating the information into the central nervous system and reacting to the musculoskeletal system and the support action of the musculoskeletal system. Motor function, including coordination, motor, and vision, vestibular sense, and sensory function, including proprioception, should act in an integrated way. However, more than half of stroke patients have motor, sensory, cognitive, and emotional disorders for a long time. Motor and sensory disorders cause the greatest difficulty in postural control among stroke patients. Objects: The purpose of this study is to determine the effect of visual and somatosensory information on postural sway in stroke patients and carrying out a kinematic analysis using a tri-axial accelerometer and a quantitative assessment. Methods: Thirty-four subjects posed four stance condition was accepted various sensory information for counterbalance. This experiment referred to the computerized dynamic posturography assessments and was redesigned four condition blocking visual and somatosensory information. To measure the postural sway of the subjects' trunk, a wireless tri-axial accelerometer was used by signal vector magnitude value. Ony-way measure analysis of variance was performed among four condition. Results: There were significant differences when somatosensory information input blocked (p<.05). Conclusion: The sensory significantly affecting the balance ability of stroke patients is somatosensory, and the amount of actual movement of the trunk could be objectively compared and analyzed through quantitative figures using a tri-axial accelerometer for balance ability.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2010년도 춘계학술발표대회
• /
• pp.289-292
• /
• 2010

Recently human activity recognition using accelerometer has become a prominent research area in proactive computing. In this paper, we present a real-time activity recognition system using a single tri-axial accelerometer. Our system recognizes four primary daily human activities: namely walking, going upstairs, going downstairs, and sitting. The system also computes extra information from the recognized activities such as number of steps, energy expenditure, activity duration, etc. Finally, all generated information is stored in a database as daily log.

• 센서학회지
• /
• 제26권1호
• /
• pp.7-14
• /
• 2017

A threshold-based fall recognition algorithm using a tri-axial accelerometer and a bi-axial gyroscope mounted on the skin above the upper sternum was proposed to recognize fall-like activities of daily living (ADL) events. The output signals from the tri-axial accelerometer and bi-axial gyroscope were obtained during eight falls and eleven ADL action sequences. The thresholds of signal vector magnitude (SVM_Acc), angular velocity (${\omega}_{res}$), and angular variation (${\theta}_{res}$) were calculated using MATLAB. When the measured values of SVM_Acc, ${\omega}_{res}$, and ${\theta}_{res}$ were compared to the threshold values (TH1, TH2, and TH3), fall-like ADL events could be distinguished from a fall. When SVM_Acc was larger than 2.5 g (TH1), ${\omega}_{res}$ was larger than 1.75 rad/s (TH2), and ${\theta}_{res}$ was larger than 0.385 rad (TH3), eight falls and eleven ADL action sequences were recognized as falls. When at least one of these three conditions was not satisfied, the action sequences were recognized as ADL. Fall-like ADL events such as jogging and jumping up (or down) have posed a problem in distinguishing ADL events from an actual fall. When the measured values of SVM_Acc, ${\omega}_{res}$, and ${\theta}_{res}$ were applied to the sequential processing algorithm proposed in this study, the sensitivity was determined to be 100% for the eight fall action sequences and the specificity was determined to be 100% for the eleven ADL action sequences.

• PNF and Movement
• /
• 제11권2호
• /
• pp.31-40
• /
• 2013

Purpose : The purpose of this study was to measure the center of mass in body with stroke patients using a tri-axial accelerometer during walking. Methods : Twenty-eight patients were recruited and divided into two groups for this study. To measure their walking ability, Timed Up & Go (TUG) test and Fucntioanl Gait Assessment (FGA) were conducted and acceleration at rotation of center of mass (COM) in body were measure for each group. Results : In the comparisons between the two groups, the TUG and FGA were not significant differences and acceleration at rotation of COM was not significant differences also. Conclusion : Our research results suggesting that the accelerometer may be used as a testing tool and ongoing assessment tool for stroke patients during effects of intervention in walking.

• 정보처리학회논문지D
• /
• 제18D권2호
• /
• pp.143-148
• /
• 2011

본 논문에서는 3축 가속도 센서를 소형 디바이스(활동량 측정기)로 구성하고 이를 사람의 신체에 착용하고 사람이 보행 시 발생하는 가속도 센서의 Raw 데이터 출력 값을 획득하여 실시간 활동량으로 변환하고 모니터링 할 수 있는 활동량 측정기와 알고리즘을 개발하였다. 피험자 59명을 대상으로 트레드밀(Treadmill)에서 호흡가스대사분석기(K4B2), Actical 그리고 본 연구에서 개발된 활동량 측정기를 착용 후 36분 동안 테스트 프로토콜에 따라 다양한 속력의 걸음(느리게 걷기, 걷기, 빠르게 걷기, 천천히 뛰기, 뛰기, 빠르게 뛰기)에 대해서 실험을 하였다. 가속도 센서의 출력 데이터와 피험자 정보를 이용하여 에너지소비량(Energy Expenditure :EE)을 추정하는 회귀식을 도출하였으며 이는 실험시 같이 착용한 Actical보다 제안하는 활동량 변환 알고리즘의 성능이 1.61% 향상 되었다.

• 한국운동역학회지
• /
• 제18권2호
• /
• pp.59-64
• /
• 2008

본 연구는 3축 가속도 센서를 허리에 부착하여 실시간으로 스포츠 동작분류를 할 수 있는 모니터 링에 관한 연구를 실시하였다. 이 모니터링 시스템은 스포츠 동작의 기본이라고 할 수 있는 걷기, 달리기, 자세변화 동지 정지상태의 동작들과 추가적으로 사이클링 동작을 분류할 수 있도록 하였다. 또한 운동 시에 발생할 수 있는 낙상을 감지하여 위급상황에 대한 정보도 나타나게 하였다. 가속도센서모듈은 인체에 부착된 형태로 스포츠 활동을 모니터링하기 위하여 소형으로 설계되었으며 활동에 방해가 되지 않게 허리에 부착되었다. 측정된 데이터는 RF통신을 통해 PC로 전송되며 알고리즘을 통해 실시간으로 동작분류를 시행하게 된다. 개발된 알고리즘을 검증하기 위한 실험으로 5명의 피험자를 대상으로 서로 다른 속도의 걷기, 달리기, 사이클링 동작을 각각 100초간 실시하였으며 낙상과 자세변화 동작(앉았다 일어서기, 누웠다 일어서기, 서있다 앉기, 누웠다 앉기, 서있다 눕기, 앉았다 눕기)은 각각 20회씩 실행하였다. 그 결과 동작분류 정확도는 95.4%를 나타내었다. 이번 연구에서 스포츠 모니터링을 통하여 정확한 자신의 운동 정보를 알려주고 운동 시에 발생하는 낙상에 대한 위급상황을 알려줌으로써 스포츠 활동에 도움을 주고자 하였으며, 추가적인 연구로 각각의 스포츠 활동에 대한 정확한 에너지 소비 추정 알고리즘을 개발 중에 있다.

• 한국콘텐츠학회논문지
• /
• 제11권12호
• /
• pp.103-111
• /
• 2011

최근 만성질환을 예방하고 건강을 증진시킬 목적으로 신체활동에 대한 중요성이 인식되면서 신체활동 연구가 활발히 진행되고 있다. 본 연구에서는 3축 가속도 동작감지기 x, y, z축에 대한 $cm/s^2$의 가속도 합인 SVM(Signal Vector Magnitude)를 이용하여 신체활동 에너지 소비량 알고리즘을 구현하였다. 기존 실험을 통해 타당도가 입증된 COUNT 방식의 Freedson, Hendelman, Leenders, Yngve 알고리즘에 SVM 방식을 적용하여 구현 하였다. COUNT와 SVM 상관관계 분석을 위하여 총 10명의(성인 남성 5명, 여성 5명, 20 ~ 30 대) 피험자를 대상으로 실험을 진행하였다. 피험자는 트레드밀위에서 3단계 신체활동 (걷기: 3km/h, 빨리 걷기: 5km/h, 러닝: 8km/h)을 1주 간격으로 4주 간 반복 실험을 진행하였다. 실험결과 얻어진 COUNT와 SVM의 간의 상관관계를 분석하여 다양한 신체활동에 따른 맞춤형 에너지 측정 알고리즘을 구현하였다.

• 재활복지공학회논문지
• /
• 제5권1호
• /
• pp.27-33
• /
• 2011

3축 가속도 동작감지기를 이용하여 x, y, z축에 대한 가속도 합인 SVM(Signal Vector Magnitude)를 적용한 신체활동 에너지 소비량과 신체활동 강도 예측 알고리즘을 구현하였다. 신체 건강한 20~30대 성인 남성 5명, 여성 5명을 대상으로 골반 장골능에 엑티그라프(LLC, USA)와 피트미터(Fit.life. korea)를 착용하고 트레드밀위에서 3단계 신체활동(걷기: 3km/h, 빨리 걷기: 5km/h, 러닝: 8km/h)을 수행하였다. 각 신체활동 단계별로 7분간 신체활동을 수행하고 5분간 휴식을 통하여 각 신체활동 단계별로 안정화된 상태에서 실험하였다. 이러한 실험을 1주 간격으로 4주간 반복 실험을 진행하였다. 실험결과 얻어진 엑티그라프와 피트미터의 600여개 데이터 상관관계를 분석하여 METs와 kcal 그리고 신체활동 강도를 구분하는 알고리즘을 구현하였다.