Recently, a novel optical player tracking system has been developed to determine positional information of athletes in a non-invasive manner. The aim of this study was to measure the accuracy and reliability of displacement estimates derived from the system. Participants completed five soccer-specific running courses at three different speeds three times each, while being filmed using the multi-camera system. The participant’s x,y field positions were determined by the optical player tracking system and displacement was estimated using Euclidean distance, and compared with real distance. On average, the difference between actual distance and estimated displacement was 0.25% (mean absolute difference = 0.79 ± 0.56m) with a slightly larger coefficient of variation during 90° turns (4.89%) when compared with straight line running (4.09%). In addition, there were strong correlations between actual distance and measured displacement (r = 0.986 – 0.988). Collectively, the typical error (0.25 – 0.36 m), typical error as a coefficient of variation (1.06 – 1.75%) and intraclass correlation coefficient (0.88 – 0.93) showed high levels of intra-operator reliability. The optical player tracking system provides accurate and reliable estimates of displacement of players on a soccer field. This system provides non-invasive position detections for players and opposition players during soccer matches.
|Number of pages||10|
|Journal||International Journal of Computer Science in Sport|
|Publication status||Published - 29 Dec 2017|