老年人跌倒检测系统研究与设计

Research and Design of Fall Detection System for the Elderly

为减少跌倒对老年人造成的身心伤害,设计了一套跌倒检测系统.将跌倒检测装置佩戴在老年人腰间,传感器可以实时采集老年人日常活动数据,检测系统根据采集到的数据通过阈值分类和支持向量机分类器双重检测识别老年人是否跌倒.该系统检测算法在设计时综合了阈值分类检测的高效性和支持向量机分类检测的准确性.一旦老年人跌倒,检测系统能够第一时间发送求救信号给120急救中心及监护人.实验表明,当有跌倒行为发生时,系统能及时报警,准确率达到预期效果.

In order to reduce the physical and mental injury of the elderly caused by falls,a set of fall detection system was designed. When the elderly wear the system detection device on the waist,the sensor can collect real-time data of the elderly′s daily activities. According to the collected data,the detection system identifies whether the elderly fall through the dual detection of threshold classification and support vector machine classifier. The detection algorithm of this system integrates the high efficiency of threshold classification detection and the accuracy of support vector machine classification detection. In case of falling,send a distress signal to 120 emergency center and family members in the first time. The experiment shows that when the fall occurs,the alarm can be timely,and the accuracy rate reaches the desired effect.