Assessment the Speed-up Detection System for Testing Sprinters’ Performances Using Computer Vision Techniques

Abstract

Objectives: This paper aims to implement the speed-up detection system for testing sprinters’ performances and also, the system is to assess with the standard video analysis software. Methods/Statistical analysis: This cross-sectional study investigates the relationship between the proposed system and the standard video analysis software (Tracker). Seven athletes (age between 19 to 42) were randomly selected for data collecting. The participants performed a fastest running with static starts at about 25 m. from the Designated point. Descriptive statistics, Cross-tabulation and also Spearman's Rho were used for statistical analysis. Besides, the root mean square error (RMS Error) and the relative root mean square error (Relative RMS Error) were employed for comparing of both systems. Findings: The results indicated the strong significant relationship between both systems at Spearman's Rho Correlation > 0.99 at p-value < .0001. The findings, moreover, revealed that the proposed system could be used to evaluate sprinters’ performances accurately. Application/Improvements: the proposed system will be useful in several fields such as biomechanics of sport, physical therapy, rehabilitation, etc.