Analysis and synthesis of function data of human movement

Abstract

The article proposes a human motion capture method based on operational data. The thesis first uses the human body wear system to perform functional processing on the captured periodic motion data, and then extracts the data sequence for the few motions. Thereafter, the classification of the vector calculation method is carried out according to the characteristics of periodic data. Through experimental research, it is found that the functional data analysis (FDA) algorithm proposed in the thesis can accurately identify human motion behaviour, and the automatically collected data has a recognition rate that is as high as 98.9%. Therefore, we have concluded that the human body data functional analysis algorithm has higher recognition accuracy than the traditional optical capture system. Thus, it is worthy of further research and discussion.

An Integrated Approach for Accurate 3D Point Cloud Recognition of Substation Equipment Using PSO and kNN Classification

“…The following is the procedure for implementing an algorithm: 1) Set the particle's initial position and speed. 2) Update the particle's position and velocity in accordance with Eqs (2), ( 3) and ( 4), [14].…”