Real-time elderly activity monitoring system based on a tri-axial accelerometer
Real-time elderly activity monitoring system based on a tri-axial accelerometer
复制标题
DOI：
10.3109/17483101003718112
复制
发表时间：
2010-01-01
影响因子：
2.2
通讯作者：
Tack, Gye Rae
中科院分区：
文献类型：
--
作者：
Kang, Dong Won;Choi, Jin Seung;Tack, Gye Rae
目的。本研究的目的是利用单腰式三轴加速度计开发一套适用于老年人的人体运动自动分类系统。提出了一种基于层次二叉树的实时运动分类算法，将日常生活活动分为四种状态：(1)坐、卧、立等休息状态；(2)行走、跑步等运动状态；(3)摔倒等紧急状态；(4)坐-立、立-坐、立-卧、卧-立、坐-卧、卧-坐、坐-卧等过渡状态。为了验证该算法的有效性，对5名健康青年受试者进行了跌倒、行走、奔跑等多种活动的实验。实验结果表明，该系统对所有活动的检测成功率约为96%。为了评估长期监测，对一名健康受试者进行了3h的家庭环境实验，98%的动作被成功分类。实验结果表明，该系统可以对日常生活活动进行监测和分类。为了进一步完善系统，有必要加入更详细的分类算法来区分几个日常活动。
Purpose. The purpose of this study is to develop an automatic human movement classification system for the elderly using single waist-mounted tri-axial accelerometer.Methods. Real-time movement classification algorithm was developed using a hierarchical binary tree, which can classify activities of daily living into four general states: (1) resting state such as sitting, lying, and standing; (2) locomotion state such as walking and running; (3) emergency state such as fall and (4) transition state such as sit to stand, stand to sit, stand to lie, lie to stand, sit to lie, and lie to sit. To evaluate the proposed algorithm, experiments were performed on five healthy young subjects with several activities, such as falls, walking, running, etc.Results. The results of experiment showed that successful detection rate of the system for all activities were about 96%. To evaluate long-term monitoring, 3 h experiment in home environment was performed on one healthy subject and 98% of the movement was successfully classified.Conclusions. The results of experiment showed a possible use of this system which can monitor and classify the activities of daily living. For further improvement of the system, it is necessary to include more detailed classification algorithm to distinguish several daily activities.