Analysis of Badminton Track Using Computer Vision Techniques

• Published in: 2023 International Conference on New Frontiers in Communication, Automation, Management and Security (ICCAMS) Vol. 1; pp. 1 - 5
• Main Authors: Yadav, Puneet Kumar, Tiwari, Shweta, Judeson Antony Kovilpillai, J, Chinnaiyan, R.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 27-10-2023
• Subjects: Aerodynamic Model; Aerodynamics; Atmospheric modeling; Badminton; Computer Vision; Convolution Neural Network(CNN); Filtering; Machine Learning; Predictive models; Real-time systems; Tracking; Trajectory
• DOI: 10.1109/ICCAMS60113.2023.10526001

For both young and old, badminton is a sport. A badminton robot must be created for the real-time dynamic tracking of badminton travelling at high speed in order to swiftly and accurately capture the dynamic route of the game. We examine the impact of gravity and aerodynamic forces on badminton flight in the air in this essay. Experiments are used to analyze the key elements influencing the badminton shuttle's flight path. A steady-state flight model for badminton is created using the badminton flight trajectory recorded by a high-speed camera, along with a spin model. Considering that both model and measurement mistakes can occur during the badminton trajectory prediction process particle filtering is an inference model based on Monte Carlo sampling that is more suited to badminton state estimation since it does not require a priori knowledge of state transfer. Therefore, in this study, particle filtering is used to follow the badminton player's path. Following experimental validation, the algorithm is quite good at following the badminton ball's motion route.