基于三轴加速度传感器的跌倒检测研究Triaxial-现代生物医学进展.PDFVIP

现代生物医学进展 Progress in Modern Biomedicine Vol.14 NO.18 JUN.2014 3585 · · doi: 10.13241/ki.pmb.20 14.18.048 基于三轴加速度传感器的跌倒 测研究* 张军建 赵 捷△ 安佰京 尹文枫 陈甜甜 李大鹏 张春游 （山东师范大学物理与电子科学学院 山东济南250014） 摘要：本文基于MMA7260QT 加速度传 器获取的人体运动加速度信号，采用人体加速度向量幅值(SVM)和人体加速度向量区 域值(SMA)描述了老年人的运动状态，检测人体跌倒，具有良好的准确性和实时性。采用bior3.3 小波分析，在轮廓的基础上，最大 程度上保留了细节，有效的去除噪声对特征量的干扰。本文提出了人体跌倒检测算法，大大降低了误 率和漏 率。首先，检测人 SVM ADL) SMA 体 是否超过阈值进行第一级跌倒检测，区别出人体日常活动（ 和跌倒；其次在此基础上，检测第一级各个跌倒的 值，是否超过阈值， 断跌倒和疑似跌倒。当两次 断都检测到跌倒发生时，报警。 ： ；小波 关键词 跌倒检测；三轴加速度 R3 18.6 TP3 12 A 1673-6273 20 14 18-3585-04 中图分类号： ； 文献标识码： 文章编号： （ ） Triaxial Accelerometer-Based Fall Detection Research* This article is based on human movement acceleration signal of acceleration sensors MMA7260QT, describing the movement state of the old with the body acceleration vector amplitude value vectors (SVM) and the body acceleration area (SMA) to detect mans fall, which has a good accuracy and real-time performance. Adopting the method of Bior3.3 wavelet analysis ,this paper effectively remove noise interference with characteristics and retained the maximum details of outline. Fall in the human body detection algorithm proposed in this paper greatly reduce the misjudgment rate and false negative rate. In order to distinguish human daily activities (ADL) and fall,the first level fall detection is to judge whether the SVM is more than the detection threshold. On this basis, this paper detected whether the SMA of the falls in the first level is more than threshold, to distinguish fall down and suspected. When the above two steps detection both judge that fall occurs, the system alerts.