Font Size: a A A

Research And Application Of Fall Detection Technology Based On Acceleration Sensor

Posted on:2020-07-05Degree:MasterType:Thesis
Country:ChinaCandidate:Y Q XuFull Text:PDF
GTID:2428330590995783Subject:Software engineering
Abstract/Summary:PDF Full Text Request
At present,the elderly population in China is increasing and has entered the aging society.The problems concerning the quality of life and physical and mental health of the elderly are increasingly prominent.According to the analysis of relevant data,the most common cause of death among the elderly is falling.It is particularly important to take active and effective measures to ensure timely and effective assistance for the elderly when they fall down.The main content of this paper is to design a fall rescue system with high performance and accuracy.Based on wearable technology,the posture analysis of human body movement is carried out according to the raw data output by the acceleration sensor.In order to improve the accuracy of the algorithm,the algorithm based on pattern recognition is improved.The overall scheme design of the system includes the hardware part and the software part.The original triaxial acceleration signal and triaxial angular velocity signal of human motion are collected by the acceleration sensor,and the original data signal is preprocessed and then analyzed.The human body common behavior movement may divide into the normal behavior movement,the violent behavior movement,the fall behavior movement three kinds big.Main research content of this paper will fall behavior movement distinguish from these three behavior movement,in order to improve the precision of fall detection algorithm,select three characteristics: three axis acceleration peak value(RA),three axis acceleration change amplitude(RS),energy(E),made up of these three characteristic values eigenvector,through a large number of data samples training,get different behavior motion corresponding to the behavior of the space,to distinguish the motion.The main functions of the fall detection system device designed in this paper are as follows.According to the measured values of the acceleration sensor,the corresponding algorithm is used to analyze the body posture of the elderly.When the falling state occurs,the microcontroller can be triggered for pre-fall treatment.Considering the serious can not handle and dispose of old people fall and fall device misjudgment,trigger micro controller for the fall,after processing the fall rescue device first sent voice distress signal into the surrounding environment,the elderly can choose whether to call the police,according to their own physical condition when the elderly unable to control their behavior,fall relief device automatic alarm.After the alarm is implemented,the fall assistance device sends the spatial location information of the elderly's falls to the preset mobile phones of relatives to ensure timely and effective rescue.
Keywords/Search Tags:Acceleration sensors, fall detection, pattern recognition
PDF Full Text Request
Related items