用户名: 密码: 验证码:
基于滑动窗口异常数据提取的跌倒行为监测方法
详细信息    查看官网全文
摘要
为了提高跌倒行为监测的准确率,减少突然下蹲等四种疑似跌倒行为造成的误报率,提出了一种基于滑动窗口异常数据提取的跌倒行为监测方法。该方法使用均值滤波法对原始三轴加速度数据进行降噪处理,为了消除手机方向性对跌倒监测的影响,将三轴加速度进行合成。利用滑动窗口技术为合成加速度数据生成一个窗口特征向量,通过计算相邻窗口特征向量间的相关系数来提取异常数据。利用决策树分类器对异常行为数据样本集进行分类,能够有效区分疑似跌倒和跌倒行为。该方法的跌倒行为识别的准确率为95%,误报率为9.5%。
In order to increase the accuracy in fall activity recognition and reduce the rate of false recognition,we put forward a kind of fall behavior recognition method based on abnoraial data extraction with sliding window.The method reduced the impact noise effects by using average filtering method.In order to eliminate impact from mobile direction,the method synthesized three axis acceleration.Using sliding window technique for the synthesis of acceleration data generated a window feature vector.By calculating correlation coefficient between the adjacent windows feature vector data to determine whether the data is abnormal.Using the decision tree classifier to classify abnormal behavior data sample set and the method can distinguish the suspected fall activity and fall activity effectively.The accuracy of falling recognition is 95%.,and the rate of false recognition is 9.5%
引文
[1]Bruno Aguiar,Tiago Rocha.Accelerometer-based Activity Recognition on Smartphone[J].2012,2-5
    [2]Reddy S,Burke J,Estrin D,etal.Determining transportation mode on mobile phones.Wearable Computers,2008.ISWC 2008.12th IEEE International Symposium on.IEEE,2008:25-28.
    [3]Yang J,Wang S,Chen N,etal.Wearable accelerometer based extendable activity recognition system.Robotics and Automation(ICRA),2010 IEEE International Conference on.IEEE,2010:3641-3647.
    [4]范琳,王忠民.穿戴位置无关的手机用户行为识别模型[J].计算机应用研究,2014,2-4
    [5]王忠民,范琳.基于蚁群算法的行为识别特征优选方法[J].西安邮电大学学报,2014,19(1):73-76
    [6]周敏,陈益强.基于移动终端的跌倒监测方法研究[D].湘潭大学,2013,18-36
    [7]Salva A,Bolibar I,Pera G,Arias C.Incidence and consequences of falls among elderly people living in the community[J].2004.172-176.
    [8]葛怡然.光电脉冲信号的时域和频域分析与测量方法研究[D].吉林大学,2010,30-36.
    [9]吴建华.中值滤波与均值滤波的去噪性能比较[J].南昌大学学报,1998,2-5.
    [10]王科俊.高效均值滤波算法[J].计算机应用研究.2010,2-5.
    [11]谢徵.基于决策树支持向量机的猪只姿态分类与异常行为分析[D].太原理工大学,2015,21-59.
    [12]刘庆和.一种基于信息增益的特征优化选择方法[D].计算机工程与应用,2011,2-5.
    [13]黄毅,佟晓光.中国人口老龄化现状分析[J].中国老年学杂志,2012,21

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700