Application of Segmentation and Tracking Algorithm in Basketball Video Image Processing
Download as PDF
Ang Li, Baoyu Zheng, Lei Li
Segmentation and tracking are always hot issues in video image processing. Aiming at the fierce movement of basketball video, a real-time segmentation and tracking algorithm is implemented. First of all, the general partition method is different. The real-time partition and background update of the two levels (pixel level and frame level) are based on the background subtraction method, and the player partition is implemented. Please use the core based target tracking method to identify team members of athletes, and then achieve accurate tracking of athletes in the fierce competition. The proposed algorithm shows that it can fully identify the team problems of each player, track multiple players without complete blocking, and track stably when the shape of player changes greatly.
Segmentation, Tracking, Background Subtraction, Video