An image recognition system for the measurement of soccer ball spin characteristics

摘要

In soccer, the world's most popular ball game, there are no systems currently in use capable of measuring soccer ball speed and spin. This paper describes a novel image recognition system capable of obtaining accurate measurements of ball velocity, elevation, spin rate and spin axis. Two sequential images of a soccer ball, a fixed time interval apart, are obtained and a series of image-processing techniques are employed to identify the spatial position of the ball within the image frame and the specific ball orientation. The soccer ball is marked with a total surface-marking scheme using coloured panels, such that a single 2D view of the ball gives a unique pattern for any orientation. A pattern matching algorithm is used to determine the specific ball orientation, and comparison of the two ball images allows the determination of ball velocity, launch elevation, spin rate and spin axis orientation to a high degree of accuracy.