基于智能手机的人体跌倒检测系统设计  

Design of human fall detection system based on smartphone

作　　者：吴志强[1] 曹蕾[1] 王凯[1] 吕庆文[1] 

机构地区：[1]南方医科大学生物医学工程学院,广东广州510515

出　　处：《计算机工程与设计》2014年第4期1465-1470,共6页Computer Engineering and Design

基　　金：广东省教育部产学研基金项目(2011B090400037)

摘　　要：为了提供老年人跌倒损伤后的紧急救助,设计了一种基于智能手机内置加速度传感器和陀螺仪的人体跌倒检测系统。阐述了利用智能手机监测人体运动变化和基于Android平台的软件设计,提出了基于信号向量模和特征量W相结合的跌倒检测算法。当检测到跌倒的发生,系统将自动发送包括用户跌倒位置、时间等紧急救助信息给预设联系人,使用户及时得到医疗救助。实验结果表明,该系统能有效检测到跌倒事件,检测结果的假阳性和假阴性较低。A fall detection system based on both accelerometer and gyroscope in smartphone is designed to rapidly response to the elders injured emergencies after the fall. Smartphone used to monitor body movement and software design based on the Android platform is described. A novel detection algorithm based on the magnitude of signal vector and the characteristics of W is pro- posed. Emergent rescue messages including fall location, time, etc. are sent to pre-selected contacts automatically to make the user receive medical treatment timely when a fall is detected. The experimental results demonstrate that fall detection system can detect the occurrence of fall events effectively and both false positives and false negatives of the detection results are low.

关 键 词：跌倒检测 智能手机 加速度传感器 陀螺仪 检测算法 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]