摘要
依托智能硬件技术和边缘计算方法,设计出一种智能跌倒检测算法。一方面内置加速度传感器,以一定频率采集佩戴者的加速度数据,并结合数据特点,基于中心矩优化模型改进三阶中心矩和四阶中心矩。另一方面,为了有效排除正常行走、上下楼梯、轻微碰撞等非跌倒情况,算法基于改进后的中心矩统计模型,引入一定长度的时间窗并综合若干判定条件进行优化,从而提升检测算法的抗干扰能力,进一步提高跌倒检测准确度。
This paper designs an intelligent fall detection algorithm based on intelligent hardware technology and edge computing method.On the one hand,the built-in acceleration sensor is inserted to collect the wearer s acceleration data at a certain frequency,and combines the data characteristics to improve the third-order central moment and the fourth-order central moment based on the central moment optimization model.On the other hand,in order to effectively exclude non-fall situations such as normal walking,going up and down stairs,and minor collisions,this algorithm relies on an improved central moment statistical model,introduces a certain length of time window and optimizes it by integrating several judgment conditions,thereby enhancing the anti-interference ability of the detection algorithm and further improving the accuracy of fall detection.
作者
解玉芳
许水燕
龚巧娴
XIE Yufang;XU Shuiyan;GONG Qiaoxian(School of Information Engineering,Xiamen Ocean Vocational College,Xiamen,Fujian 361100,China;College of Artificial Intelligence Xiamen Institute of Technology,Xiamen,Fujian 361021,China)
出处
《福建技术师范学院学报》
2025年第5期20-31,共12页
JOURNAL OF FUJIAN POLYTECHNIC NORMAL UNIVERSITY
基金
厦门市自然科学基金项目(3502Z202374074)。
关键词
跌倒检测
中心矩
加速度传感器
报警系统
fall detection
central moment
acceleration sensor
alarm system
