Sports Video Annotation and Multi-Target Tracking using Extended Gaussian Mixture model

Video offers solutions to many of the traditional problems with coach, trainer, commenter, umpires and other security issues modern team games. This paper presents a novel framework perform player identification tracking technique for sports (Kabaddi) extending implementation towards event handling process which expands game analysis third umpire assessment. In proposed methodology, video preprocessing has done Kalman Filtering (KF) technique. Extended Gaussian Mixture Model (EGMM) implemented detect object occlusions labeling. Morphological operations have given more genuine results on detection spatial domain by applying silhouette spot model. Team localization Robust Color Table (RCT) model generation classify each members. Hough Grid Transformation (HGT) Region Interest (RoI) method applied background annotation process. Through court line tracing labeling in half respect their state-of-art foremost is performed. Extensive experiments been conducted real time samples meet out all challenging aspects. Proposed algorithm tested both Self Developed (SDV) data Real Time (RTV) dynamic greater accuracy performance measures different state samples.