Paper Key : IRJ************576
Author: Aditya Prakash Rai,Harsh Maheshwari,Deepanjal Uppal
Date Published: 19 Nov 2024
Abstract
This project aims to build a robust, automated football analytics system that leverages machine learning and computer vision to produce real-time, data-driven insights into both individual and team performance. By processing video footage from football matches, the system can detect, identify, and track players, referees, and the ball with precision. Using advanced object detection and tracking algorithms, it ensures that the identified entities are continuously monitored throughout the game, providing seamless transitions between frames. A custom-trained model further optimizes detection, accommodating the highly dynamic and variable nature of football gameplay. To enhance usability, team classification algorithms are employed to distinguish players by team, simplifying data interpretation and allowing for targeted analysis. Key metrics such as player speed, distance covered, and ball possession are comprehensive calculated, assessment offering of a in-game performance. This data can be invaluable for coaches, analysts, and players, as it helps them gain insights into strategic movements, player positioning, and overall team dynamics. The system's end goal is to provide a reliable, automated solution that enhances game strategy through real-time analysis, thereby supporting informed decision-making and performance improvement in competitive football scenarios.
DOI LINK : 10.56726/IRJMETS64054 https://www.doi.org/10.56726/IRJMETS64054