The Design And Implementation Of Human Posture Monitoring System Based On Accelerometer

Human posture monitoring system is a decision of human posture using accelerometer tapped on human body. This research is important to monitoring human movement and posture under home environment and automatic aid system. For example, it can be obtain through human posture monitoring system that current human body movement, possible fall, and it can indirectly evaluate the human activity ability, measure functional ability and level of human activity about human body. Thus, with the growing of living standard, human posture monitoring system is become a hot research area about intelligent life.Under this research background, this paper do some research from the following three aspects. Fitst, this paper analysed the acceleration characteristic about human movement, and found how to select the appropriate accelerometer. Then this paper analysed existing algorithms about human movement classification, proposed a classification model based on binary decision tree, and implemented a classification algorithm based on binary decision tree model. At last, this paper designed and implemented human posture monitoring system based on accelerometer, and designed experiments to test the system.Through experiment, the result indicates the accuracy of the system can achieve97.02%generally. The distinguish of static posture can achieve98.3%, the detection accuracy of fall can achieve98%. This demonstate human posture monitoring system based on accelerometer can provide high accuracy, and is very feasibility.