The traditional basketball training is unable to be quantified and shared with others for it heavily relies on the coach's experience. In this paper, we develop an inertial measurement unit (IMU) to collect movement data of the basketball player and identify his postures, which helps improve the coach's guidance and the athletes' skills in a quality manner. Additionally, the IMU sensor is designed to recognize nine kinds of basic basketball movements, such as stand, walk, run, jump, in-situ dribble, dribble while walking, dribble while running, set shot, and jump shot. Experimentally, the IMU sensor is worn on the player's right wrist. Meanwhile, the player's movements are captured by the sports camera (GoPro Hero 6) for reference. Further, five features are extracted from y-axis acceleration (YAA) data and z-axis angular velocity (ZAAV) data for the analysis of basketball movements. Finally, a 98.9% accuracy rate of recognizing each basic basketball movement of one player is achieved by using a neural network algorithm.
This paper presents a novel MoG based method for foreground detection and segmentation in video surveillance. Normal MoG is different to deal with the foreground objects that stay in the scene for a long time and segment difficult foreground objects from one blob. We improve MoG by adopting posterior feedback information of Kalman filter tracking, to robustly modeling the background and to perfect the foreground segmentation result. Experiments and comparisons show that our method is robust and accurate in video surveillance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.