Implementation of a Real-time Fall Detection Algorithm Based on Body's Acceleration
Zhangqin Huang · Journal of Chinese Computer Systems · 2012
The real-time falling detection can provide old people timely alert and call for help.This paper introduces such an algorithm based on single triaxial accelerometer according to kinematic characteristics of human.We Abstract value of overweight,continuous weightless time,tilting angle,still time as features,and implement a decision tree classifier running on an Android smartphone.The algorithm does not require fixing the orientation of sensor.We selected data from daily activities and a set of falling activities with similar acceleration feature for test.The results are that average response-time of the algorithm is less than 6 seconds and average accuracy is 92%,which shows its effectiveness.