Inertial Sensor-Based Approach for Shot / Pass Classification During a Soccer Match

Coaches and players in soccer are heavily interested in statistics like number of shots and passes during training sessions and competitions. Currently available systems use video or computerized technology to create such statistics and are therefore mainly applicable for elite teams. Thus, the purpose of this paper was to develop a low-cost inertial sensorbased approach for shot/pass classification in soccer, which can be used by teams at amateur level. High intensity peaks were detected in the accelerometer data from the left and right shoe of soccer players. Segmented windows around the peaks were classified regarding shot, pass, and other. The system was evaluated based on data from 12 players during a 60-minute soccer match. The system was able to differentiate between shots and passes with an overall mean classification rate of 84.2 %. The proposed approach is one important step toward a detailed soccer match analysis system, which provides coaches and players of amateur teams with important statistics for performance assessment.