Basketball Court Detection Based on Adaptive Gaussian Mixture Model and the Application in Basketball Video Annotation

Basketball court detection is the key step of video annotation, however the court present different color when using different camera or in different play court. In this paper the adaptive Gaussian Mixture Model (GMM) is used to build the model of the basketball court, which is used in the basketball video annotation. The proposed algorithm extracts the main color of the court´s main color using the GMM and IEM. Under these conditions the basketball video is annotated into in-play and close-up. Experimental results show that the proposed algorithms are effective.