Basketball motion video target tracking algorithm based on improved gray neural network

This article takes the basketball game video with high attention in sports video as an example to analyze the feature extraction of basketball game video, improve the gray neural network algorithm, and disassemble the basketball video. Moreover, this paper takes basketball, basket, and athletes as the feature extraction objects. Considering that the basketball is a round sphere and the object in the image is a circle, as well as the edge is added to the original image and saved. In addition, this paper combines the improved gray neural network algorithm to construct a basketball motion video target tracking algorithm. Finally, this paper designs experiments to verify the performance of this method. The experimental test results show that this method can effectively recognize basketball gestures with high recognition accuracy, which provides a new method for basketball posture recognition.