Method for analyzing skiing motion sequence based on hidden Markov analysis

Abstract

The invention discloses a method for analyzing a skiing motion sequence based on hidden Markov, and belongs to the field of motion attitude data processing and analysis. The implementation method comprises the following steps: windowing original attitude data of a skier acquired by a sensor, dividing the original attitude data into data frames on a time sequence, and extracting a'symbolic value 'representing an attitude matrix by utilizing SVD and a matrix norm; building a probability transfer model of the skiing motion sequence based on hidden Markov, optimizing the probability transfer modelparameters of the skiing motion sequence by using the'symbolic value 'sequence on the time sequence, and calculating the optimal path of the hidden state of the skiing motion posture sequence. The probability of each skiing state at the next moment is calculated through the optimal hidden state path and the probability transfer model, then the skiing state at the next moment is predicted, the motion state of a skier can be evaluated according to the motion posture prediction data of skiing of the skier, and then the training effect on a skiing trainer is improved.

Description

technical field [0001] The invention relates to a method for analyzing skiing motion sequences based on Hidden Markov, and belongs to the field of motion posture data processing and analysis. Background technique [0002] The 24th Winter Olympic Games will be held in my country in 2022, and people's enthusiasm for ice and snow sports continues to rise. Whether it is an amateur or a professional athlete, it is necessary to obtain the action sequence during the skiing process during training, so as to analyze and guide their skiing process. The most important technology in the field of obtaining human motion posture sequences is motion capture technology. Many scholars at home and abroad have conducted a lot of research on obtaining human body motion posture sequences using existing motion capture technologies. The current mainstream motion capture methods are divided into There are three kinds of photoelectric capture, image capture and inertial sensor capture. In the field...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a kind of method based on Hidden Markov analysis ski motion sequence, can solve following technical problem: (1) reduce the dimension of attitude matrix by using SVD singular value decomposition and matrix norm, and then pass low dimension The data represents the attitude information, reducing the amount of calculation; (2) building the probability transition model of the skiing sequence with the help of the hidden Markov method; (3) optimizing the parameters of the skiing sequence model through the observation state; (4) using the obtained skiing sequence The probability transition model analyzes the motion sequence to realize the analysis of the athlete's skiing process

Images