PLAYER POSITION DETECTION AND MOVEMENT PATTERN RECOGNITION FOR AUTOMATED TACTICAL ANALYSIS IN BADMINTON

This thesis documents the development of a comprehensive approach to automate badminton tactical analysis. First, a computer algorithm was developed to automatically track badminton players moving on a court. Next, a machine learning algorithm was developed to analyse these movements and understand their underlying tactical implications. Both algorithms were tested and validated using video footage recorded at International badminton tournaments. The results demonstrate that the combination of these two algorithms can successfully automate badminton tactical analysis.