The fitting method and system of basketball shooting curve based on neural network

Abstract

The invention provides a neural network-based basketball shooting curve fitting method and system. The method includes: training the neural network model according to the training shooting video to obtain a basketball recognition model; inputting the shooting video to be recognized into the basketball recognition model, and identifying all Basketball position; obtain the first preset number of fitting curves of the shooting video to be identified, and calculate the fitting value of each fitting curve, and use the fitting curve corresponding to the largest fitting value as the basketball shooting curve; obtain the fitting The curve and the corresponding fitting value include: randomly obtaining a second preset number of video frames to be recognized from all the video frames to be recognized in the shooting video to be recognized, and according to the acquired basketball position in the video frame to be recognized according to the parabolic calculation formula Calculate the fitting curve; calculate the degree of coincidence between the basketball position and the fitting curve of all unacquired video frames to be recognized, and use the degree of coincidence as the fitting value. Beneficial effect: the basketball shooting curve with higher accuracy is calculated.

Description

technical field [0001] The invention relates to the technical field of sports training assistance, in particular to a method and system for fitting a basketball shooting curve based on a neural network. Background technique [0002] Basketball training often requires a lot of time and energy. Through continuous trial and repeated exploration, athletes gradually form their own shooting methods and shooting habits. If there is no standardized coaching guidance in the middle, athletes often easily form wrong exercise habits, which will affect shooting accuracy. and stability. [0003] The current existing technology uses tracking the trajectory of the basketball, and compares the trajectory of the basketball with the standard fitting curve. However, the above-mentioned prior art obtains the trajectory of the basketball through tracking, but the tracking algorithm for tracking the basketball cannot be 100% accurate. All the basketballs in the basketball trajectory are detected,...

AI Technical Summary

Problems solved by technology

[0002] Basketball training often requires a lot of time and energy. Through continuous trial and repeated exploration, athletes gradually form their own shooting methods and shooting habits. If there is no standardized coaching guidance in the middle, athletes often easily form wrong exercise habits, which will affect shooting accuracy. and stability

[0003] The current existing technology uses tracking the trajectory of the basketball, and compares the trajectory of the basketball with the standard fitting curve. However, the above-mentioned prior art obtains the trajectory of the basketball through tracking, but the tracking algorithm for tracking the basketball cannot be 100% accurate. All the basketballs in the basketball trajectory are detected, so there are basketballs that are prone to missed or false detections in the basketball trajectory obtained in the prior art

Images