Tennis Assistance Technology Based on Dynamic Time Warping Algorithm

Abstract

With the improvement of economic level, people’s demand for sports activities is increasing, especially for on-net opposability sports such as tennis. However, learning tennis techniques is often difficult for beginners and requires a lot of repeated practice to master. Traditional teaching methods are inefficient and difficult to quantify the correctness of actions. In view of this research, a tennis sports assistance technology based on dynamic time warping algorithm is developed. By collecting athletes’ motion data and using dynamic time warping algorithm for motion similarity analysis, personalized technical improvement suggestions are provided for athletes. This technology combines components such as normalization, support vector machine, joint detection, sparse matrix, and second-order stepping mode to improve algorithm performance and reduce computational complexity. The experiment outcomes indicate that this method can validly raise the training effect of tennis players, with an accuracy rate of 95.66%, a calculation time of 0.32 seconds, a variance of 0.88, and an average absolute error of 4.22. Compared with the experimental group that does not use normalization, support vector mechanism node detection, sparse matrix, and second-order stepping mode, there is a significant improvement in performance. Therefore, technology significantly improves the scientific and targeted nature of tennis training through advanced algorithms and data processing techniques. This technology not only provides real-time and accurate feedback to help athletes improve their technical movements, but also enhances training productivity and precision, which is important for promoting the popularization and development of tennis.

Keywords

• DTW
• tennis sports
• sports assistance
• support vector machine
• motion detection