Extraction Method of Athlete's Shooting Motion Trajectory Based on Image Segmentation Algorithm

Abstract

The shooting action is one of the commonly used offensive methods in basketball games. This article proposes a method of extracting the basketball shooting action based on the image division algorithm. Aiming at the unique feature of the basketball shooting object in the basketball video, the video image is mixed with Gaussian modeling, and the Grab Cut algorithm is used to divide the basketball object to realize the extraction of the basketball foreground. The results of the simulation experiments show that in the video of the basketball ball, the goal of the basketball shot is divided and extracted. It has good results, and the method has the advantages of less interactive operations and high accuracy of division, which can extend the color field of the image.