Research On Human Activity Recognition And Capturing Technology Based On WBAN

As a new research hotspot in recent years,the Wireless Body Area Network(WBAN)is closely related to the health of everyone.With the development of technology and the change of time,the wireless body area network will bring great changes to human beings.Personal real-time health monitoring and disease prevention will come true in the future.Real-time monitoring of vital signs parameters can be performed through wireless body area network devices,and personal information can be archived for disease prevention and health guidance through long-term data trends.In addition to escorting human beings in medical rehabilitation,wireless body area networks also have broad application prospects in sports training,military operations and leisure and entertainment,involving all walks of life in society,so more and more domestic and foreign scholars have invested research work on wireless body area networks.Attitude capture and recognition based on body area network has always been one of the difficulties in wireless body area network research,mainly reflected in the randomness of human motion and the efficient collection and processing of body area wearable device data.At present,domestic and foreign scholars have proposed a large number of algorithms for body posture recognition based on body area network,which solves a large number of human body gesture recognition problems.However,there is still a lack of research on human body gesture recognition and capture system.The common optical motion capture systems used in film production are expensive and limited in application environments,and are not suitable for personal gesture recognition and capture.This paper studies the human activity recognition and capture technology based on WBAN.Taking the attitude recognition and capture technology as the starting point,the inertial sensor acquires and captures the human body posture information in real time.The self-built gesture recognition and capture system is the foothold.The basic attitude tracking display shows the attitude recognition and capture prototype system;the high-precision recognition algorithm based on machine learning is the innovation point,achieving high-precision recognition of no less than 5 basic human postures,for further posture judgment,state Discrimination provides a basic pose unit.The main research work of this paper is as follows:Above all,build the hardware of body activity recognition and capture based on body area network.Firstly,the overall scheme design is carried out,and then the selection and circuit design of the core control module,motion capture module and wireless communication module are carried out.At the same time,the layout and wear design of the overall hardware system are completed,and finally the hardware circuit is debugged.Furthermore,the software design of the gesture recognition and capture system based on the body area network is completed.Firstly,the overall scheme design of the software system is carried out,and the modular software design idea is adopted.Then the data acquisition and processing subroutine design and the subroutine software interface subroutine design are carried out.Finally,the overall debugging and function verification of the system are completed under the hardware platform.Last,the design of human body gesture recognition algorithm based on body area network is completed.Firstly,a hierarchical human body gesture recognition architecture is proposed,which divides the basic human body gesture recognition problem into two processes,and then uses the Particle Swarm Optimization(SSO)Optimized Support Vector Machine(SVM)classification.The gesture recognition process is completed,and finally the performance analysis results show that under the classification architecture,high-precision recognition of five human poses can be realized.In the above research,the human activity recognition and capture based on WBAN is deeply studied from the hardware,software and algorithm layers.The purpose of this paper is to explore the posture recognition and capture technology in body health monitoring and sports training and other applications.The high-precision recognition algorithm of human body posture is proposed from the theoretical aspect,and the motion capture system based on body area network is built.The real-time capture and reproduction of basic human motion is realized,which lays a foundation for the application of this technology in wireless body area network.