Football Player Detection and Tracking by Coupled Kalman Filter, Training Algorithm and Object Detection Methods

This paper describes a camera-based observation system for football games that is used for the automaticanalysis of football games and reasoning about multi-agent activity. The algorithm runs on video streamsproduced by cameras set up for TV broadcasting. Our approach consisted of read video files, convert video filesto frame, player detection using coupled object detection methods and training algorithms and tracking withKalman filter. The result showed that the accuracy of player detection and tracking of the movies with low speedwas more than high speed. Furthermore when the region of filed was nearly constant the accuracy of detectionand tracking increased, because of training algorithm effects on detection. When the players run with constantspeed, the accuracy of detection and tracking was very high until the players run acceleratory. However Kalmanfilter coupled with training algorithm and object detection methods is efficient for detecting and tracking thefootball players very well.

Keywords:

Football game , Kalman filter , player detection , player tracking , training algorithm