Intelligent badminton racket action recognition system

Abstract

The invention discloses an intelligent badminton racket action recognition system which comprises a racket comprising a racket head and a racket handle; a human body signal acquisition device, used for acquiring the three-axis acceleration, the three-axis angular velocity and the three-axis attitude angle of the racket, the human body signal acquisition device comprising a main control chip, a motion sensor, a Bluetooth module and a lithium battery, and the main control chip being electrically connected with the motion sensor, the Bluetooth module and the lithium battery; and a human body action analysis device, used for decoding the data acquired by the human body signal acquisition device and obtaining an identification result by utilizing a motion identification method, and the human body action analysis device having a network training mode and a competition mode. According to the intelligent badminton racket action recognition system, the effect of more accurate training is achieved by learning the exercise data sample of each athlete, and the starting point and the ending point of each action are determined by adopting an exercise recognition method, so that the counting function of various actions is realized.

Description

technical field [0001] The invention relates to the technical field of badminton equipment, in particular to an intelligent badminton racket action recognition system. Background technique [0002] Badminton is a popular sport that is not severely restricted by the venue. In the process of swinging the racket, the strength of the upper limbs of the human body is exercised, and in the process of running, emergency stop, jumping, and turning back and forth, the athletes are exercised. lower body strength. Badminton is relatively comprehensive for the development of the human body, and at the same time has a positive impact on the cardiopulmonary function of the human body. [0003] At present, scholars at home and abroad have carried out research on human motion recognition algorithms. Allen et al. used distributed sensor networks for human motion recognition, and proposed a linear settlement method to process sensor network data, but there are many sensors, large data volume...

AI Technical Summary

Problems solved by technology

This algorithm is more suitable for processing non-numerical data types, but not good at continuous data types, and the algorithm has poor learning ability, needs to customize different logic rules for each athlete, and the amount of calculation is relatively large

When the types of recognition increase, the rules will become more complex, which greatly increases the error rate of model recognition

Images