Mathematical Modelling & Simulation for the Qualification of Tennis Stances Improvement for Sports Player using 2D video analysis using DIP
Main Article Content
Abstract
In this paper, the simulation & qualification of the improvement of tennis stance for player performance improvement using 2d analysis of videos taken from a mobile camera is presented along with the simulation results. This research introduces an innovative approach to improving tennis performance by optimizing players' biomechanical stances during specific shots, using advanced 2D video analysis combined with Recurrent Neural Networks (RNNs). By employing precise pose estimation algorithms, the study meticulously captures skeletal keypoints to calculate joint angles using vector dot product calculations. These keypoints provide a detailed biomechanical analysis and allow for the categorization of movement patterns through unsupervised clustering techniques like k-means. The study further enhances the accuracy of these analyses by employing adaptive acceptance areas defined by various distance metrics, addressing challenges such as motion artifacts, fluctuating lighting conditions, and low signal-to-noise ratios with high-SNR imaging equipment and finely tuned camera calibration. The methodology ensures the capture of high-quality data crucial for effective computational analysis. It utilizes cloud computing to process data while ensuring data confidentiality and leveraging the scalability of computational resources. This robust integration supports detailed kinematic analysis via part affinity fields and TensorFlow Lite, facilitating immediate feedback on players’ movements and biomechanical alignment. This research significantly advances the field by integrating sophisticated computational algorithms and customized hardware solutions that go beyond the constraints of conventional video analysis. By conducting an in-depth kinematic analysis of player movements and creatively applying clustering algorithms, the study offers a thorough method for boosting tennis performance. This technique not only improves current coaching methodologies but also establishes new benchmarks in sports performance analysis, ultimately seeking to transform tennis coaching with data-driven insights and technological innovations. The effectiveness of this model, demonstrated through the research, has potential applications across various scientific and engineering fields. The simulations shows the effectiveness of the methodology that is being developed by us. The modelling results shows the effectiveness of the method developed, which could be used for a host of science & engineering applications.
Article Details

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.