Basketball tracking using mean shift algorithm

Abstract

Video analysis plays a vital role in commercial application, sports and military systems. Various methods are presented in literature. Mean shift algorithm is presented in this paper for basket ball tracking because it is more efficient than other that is defined by histograms. The tracking is the important block in the detection and recognition of the basket ball. Different object tracking algorithms are investigated. The performance of tracking in two video sequences is performed and the method gives 91.3% precision for video sequence 1 and 93.6% for sequence 2.