Sports Video Image Analysis Application Based on BP Neural Network Algorithm

Abstract

Human motion analysis is one of the main directions in the field of computer vision in recent years. Especially in the emerging field of competitive sports analysis. The purpose of this paper is to study the application of sports video image analysis based on BP neural network algorithm. The node number of input layer, output layer and hidden layer of BP neural network classifier is selected to establish the neural network classifier model. Select parameters such as learning rate and convergence error. The fusion features of LBP histogram reflecting motion behavior and motion speed are taken as the input of BP neural network, and the type of motion behavior is taken as the output of the neural network. Through experimental research, it is concluded that the sports behavior recognition algorithm proposed in this paper can achieve good results.