摘要
跌倒会对老人健康产生大的伤害,因此跌倒检测系统的重要性日益凸显。针对老人跌倒的复杂运动场景和噪声数据含有的大量有用信息,提出了一种噪声嵌入的跌倒检测系统。系统使用3D加速度传感器采集运动的加速度数据,分别对数据进行跌倒标注和噪声强度标注,然后对数据进行特征提取。使用特征和标注信息分别训练噪声强度分类和跌倒检测分类器,最终使用训练好的两个分类器实现跌倒检测。数据采集传感器设计为可穿戴设备,使用Zigbee进行组网,服务器端使用Java编写一个服务器程序,实现了对数据的分析与处理。实验结果表明系统能满足老年人日常生活中的需求,对一些意外跌倒能够给予及时的检测与报警。
Fall will be a great harm to the health of the elderly, so the importance of fall detection system has become increasingly prominent. A fall detection system based on noise embedded algorithm is proposed in this paper, since the old man fall movement is very complex and the noise data contains a large amount of useful information. The 3D acceleration sensor is used to collect the movement of accelerating data, the samples are labeled by intensity of fall and noise and extraction of feature is done on the data. The feature and label are used as train classifiers of fall detection and intensity of noise. At last the trained classifier is served to recognition of fall. The sensor of data acquisition is designed as a wearable device, using Zigbee for the network and Java for writ ing a program to process and analyze the data. The experimental results demonstrate that the system can meet the needs of the elderly in their daily lives, and some unexpected falls are able to be detected and alarmed timely.
出处
《微处理机》
2017年第2期74-76,81,共4页
Microprocessors
基金
南京航空航天大学研究生创新基地(实验室)开放基金(kfjj20150401)
关键词
噪声嵌入
跌倒检测
模式识别
支持向量机
主成份分析
特征提取
Noise Embedding
Fall Detection
Pattern Recognition
Support Vector Machine
Principal Component Analysis
Feature Extraction
