北京生物醫(yī)學工程

基于多傳感器數(shù)據(jù)融合的跌倒檢測算法

A fall detection algorithm based on multi-sensor data

作者：李坤姜萍萍顏國正

單位：上海交通大學電子信息與電氣工程學院醫(yī)學精密工程與智能儀器研究所（上海200240）

分類號：R318.04

出版年·卷·期（頁碼）：2016·35·5（483-488）

摘要：

目的跌倒在老年人生活中是一種常見的現(xiàn)象，是致使老年人發(fā)病和死亡的主要原因之一。實時的跌倒檢測系統(tǒng)能夠及時報警，縮短等待救治的時間，減少由跌倒引起的意外傷害。可是，在大多數(shù)的跌倒檢測系統(tǒng)中，人們僅利用加速度計設計檢測系統(tǒng)，基于單一數(shù)據(jù)的算法不能完整表征跌倒時身體姿態(tài)變化的信息。為此本文擬采用陀螺儀和加速度計的數(shù)據(jù)設計跌倒檢測的算法。方法首先介紹了利用MEMS慣性傳感器設計置于腰間的可穿戴的跌倒檢測系統(tǒng)，然后對跌倒的規(guī)律進行了分析，基于此提出了基于多傳感器數(shù)據(jù)融合的跌倒檢測算法，即通過數(shù)據(jù)融合的技術提取出身體加速度及其動態(tài)量和靜態(tài)量、加速度變化量、身體姿態(tài)角、角速度絕對值之和等特征參數(shù)，利用多參數(shù)設計了基于閾值判定的跌倒檢測算法。結果收集10名志愿者做模擬跌倒以及日常活動的數(shù)據(jù)，對算法的有效性進行驗證，取得96.67%的靈敏度和97%的特異性，并且此指標高于Kagans等算法的結果。結論本文提出的算法在跌倒檢測中具有較好的有效性及優(yōu)點。

Objective Fall, a common phenomenon, remains a major source of morbidity and mortality among older adults. Real-time detection system of falls can alarm in time when falling happens, shorten the waiting time for treatment to reduce injuries caused by a fall. In most of the fall detection systems, however, people only employ accelerometer to design detection systems so that these algorithms do not completely demonstrate the characteristic of falling. So this paper uses gyroscope and accelerometer sensors to devise the algorithm. Methods This study, firstly, designs a wearable fall detection system on the waist by using the MEMS inertial sensors. Secondly, according to the law of fall, this study proposes a fall detection algorithm based on multi-sensor data fusion, which extracts the body acceleration and its dynamic and static parameter, the body attitude angle, the absolute values of angular velocity. Third, a fall detection algorithm is designed based on multi-sensor data by using the parameters. Results We evaluate the algorithm on data recorded from 10 volunteers performing falls and activities of daily living (ADL), achieving 96.67% sensitivity and 97% specificity, superior to the algorithm reported by Kagans et al. Conclusions The algorithm based on multi-sensor data fusion is a more suitable method of fall-detection.

參考文獻：

［1］Kangas M, Vikman I, Wiklander J, et al. Sensitivity and specificity of fall detection in people aged 40 years and over［J］. Gait Posture, 2009, 29:571-574.

［2］Felisberto F, Moreira N, Marcelino I, et al. Elder care’s fall detection system［M］// Ambient Intelligence-Software and Applications. Berlin: Springer Berlin Heidelberg, 2011: 85-92.

［3］Bagalà F, Becker C, Cappello A, et al. Evaluation of accelerometer-based fall detection algorithms on real-world falls［J］. Plos One, 2012, 7(5): e37062.

［4］Bourke AK, Van de Ven P, Gamble M, et al. Evaluation of waist-mounted tri-axial accelerometer based fall-detection algorithms during scripted and continuous unscripted activities［J］. Journal of biomechanics, 2010, 43(15): 3051-3057.

［5］Chao PK, Chan HL, Tang FT, et al. A comparison of automatic fall detection by the cross-product and magnitude of tri-axial acceleration［J］. Physiological Measurement, 2009, 30(10): 1027.

［6］Tong L, Song Q, Ge Y, et al. HMM-based human fall detection and prediction method using tri-axial accelerometer［J］. Sensors Journal, IEEE, 2013, 13(5): 1849-1856.

［7］Rougier C, Auvinet E, Rousseau J, et al. Fall detection from depth map video sequences［M］// Toward Useful Services for Elderly and People with Disabilities. Berlin: Springer Berlin Heidelberg, 2011: 121-128.

［8］Rougier C, Meunier J, St-Arnaud A, et al. Robust video surveillance for fall detection based on human shape deformation［J］. Circuits and Systems for Video Technology, IEEE Transactions on, 2011, 21(5): 611-622.

［9］Vaidehi V, Ganapathy K, Mohan K, et al. Video based automatic fall detection in indoor environment［C］. Recent Trends in Information Technology (ICRTIT), 2011 International Conference on. IEEE, 2011: 1016-1020.

［10］Fong WT, Ong SK, Nee AYC. Methods for in-field user calibration of an inertial measurement unit without external equipment［J］. Measurement Science and Technology, 2008, 19(8): 085202.

［11］Noury N, Lundy JE, Rialle V. Processes and system for detection of abnormal situations of a person in a living space: U.S. Patent 8 223 011［P］. 2012-7-17.

［12］Schwickert L, Becker C, Lindemann U, et al. Fall detection with body-worn sensors［J］. Zeitschrift für Gerontologie und Geriatrie, 2013, 46(8): 706-719.

［13］Huang B, Tian G, Li X. A method for fast fall detection［C］. Intelligent Control and Automation, 2008. WCICA 2008. 7th World Congress on IEEE, 2008: 3619-3623.

［14］Nyan MN, Tay FEH, Murugasu E. A wearable system for pre-impact fall detection［J］. Journal of biomechanics, 2008, 41(16): 3475-3481.

服務與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請登錄！點此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频