CAJ | 학술논문

摔倒作为人体活动的一部分，是影响人体健康的一大因素，尤其对病人和老年人而言，摔倒检测至关重要。基于MEMS三轴加速度传感器采集的人体活动加速度信号，提出了一种基于固定阈值的信号幅度向量滑动平均法SVMSA。该方法根据人体活动时的加速度信号特征，利用预先设定的阈值对加速度信号幅度向量SVM的滑动平均SVMSA进行判决，同时使用差分信号幅度域DSMA区分快速跑步等剧烈运动，准确实现了人体的摔倒检测。主要优势在于分析并区别了人体快速跑步等剧烈运动对摔倒检测的影响。通过对8位实验者的测试，该算法实现了94.4%的精确度。实验表明该算法能够较为准确地实现人体的摔倒检测。
솔도작위인체활동적일부분，시영향인체건강적일대인소，우기대병인화노년인이언，솔도검측지관중요。기우MEMS삼축가속도전감기채집적인체활동가속도신호，제출료일충기우고정역치적신호폭도향량활동평균법SVMSA。해방법근거인체활동시적가속도신호특정，이용예선설정적역치대가속도신호폭도향량SVM적활동평균SVMSA진행판결，동시사용차분신호폭도역DSMA구분쾌속포보등극렬운동，준학실현료인체적솔도검측。주요우세재우분석병구별료인체쾌속포보등극렬운동대솔도검측적영향。통과대8위실험자적측시，해산법실현료94.4%적정학도。실험표명해산법능구교위준학지실현인체적솔도검측。
As a part of human activities,fall is one of the key factors affecting human health,especially for patients and elders,fall detection is of much importance. This paper presents a method of Signal Magnitude Vector Sliding Average(SVMSA)with Fixed Threshold,based on the acceleration signals of human activity acquired from a MEMS tri-axial accelerometer. By extracting the characteristics of human activity acceleration signals, this algorithm accurately achieves human's fall detection,using the prefixed threshold to judge the SVMSA and the differential signal magnitude area( DSMA) to distinguish fast running. The major advance lies in the attempt to analyze and distinguish human's fall and other intense activities,like running fast. By testing eight participants,we get 94. 4% accuracy. Ex-perimental results indicate the proposed algorithm can realize human's fall detection with much accuracy.