Introduction: Laser sensors for analyzing velocity in real-time is adequate (Ferro et al, 2014) although its use has been limited to placing them on tripods (Graubner and Nixdorf, 2011) or to guide them manually, being easy to lose the bean contact with the athletes` backs. The purpose was to analyze the reliability test re-test of a laser system (BioLaserSport®) equipped with an automatic tracking which is controlled by computer vision for monitoring the athletes in order to obtain their running velocity. Methods: Eighteen male, recreationally active healthy students (age=20.18 ± 1.56 years, height= 1.77 ± 0.05 m, body mas=69.19 ± 5.34 kg) took part in two series of 30 m sprint during two days in the same week. Distances were measured in real time by a laser system Biolasersport® (Ferro, 2012; Ferro and Floria, 2010) using a LDM301-Jenoptik-type1-laser at 2000 Hz. For to develop an algorithm that allows tracking clearly and concise of the location of the marker used by the athlete, we use the open library Open Computer Vision-open CV. It allows to apply techniques of computer vision. Calchist function, to select the region of interest (the color reference) for calculating the histogram was used. The center of region or center of mass provides x,y coordinates into the image analyzed. This coordinates were used to make the control of the positions and orientations of Biolasersport®. An integral proportional control (IP) to calculate the error between the center of the frame and center visual reference for provide the correction signal required on the system was used.. Average velocities (Vm) and maximum velocities (Vmax), over sections 0-10, 10-20 and 20-30 m were analyzed. Intraclass correlation coefficient (ICC) with confidence interval (CI) of 95%, standard error of measurement (SEM) and the minimum difference (MD) to be considered ¨real" were used to analyses Intrasession and interssesion reliability. Results: The ICC intrasession and intersession values obtained were 0.85-0.93 for Vmax and 0.85-0.97 forVm. SEM was lower than 0.14 m/s for Vmax and 0.09 m/s for Vm. MD was lower than 0.28 m/s for Vmax and 0.24 m/s for Vm. Discussion: The ICC intrassesion was excellent being higher in Vm than in Vmax and better than ICC intersession. The minimum value at which a change in performance would be considered would exceed 4.67%. The automatic control implemented in the laser system by computer vision can improve reliability in the analysis of the velocity in real time, eliminating human error produced when aiming the laser sensor manually. Also, it can be used in competition conditions without the colours or shapes the environment affect their measure.
© Copyright 2016 21st Annual Congress of the European College of Sport Science (ECSS), Vienna, 6. -9. July 2016. Veröffentlicht von University of Vienna. Alle Rechte vorbehalten. / Übersetzt mit https://www.DeepL.com/Translator