Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm
Table tennis is a highly technical sport. Beginners often struggle to correct and optimize their movements because they lack professional knowledge and guidance. To enhance the skills of beginners, a diagnosis model for table tennis beginner movement problems based on sensors and an improved dynamic...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-12-01
|
| Series: | Systems and Soft Computing |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772941925001681 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850117316818239488 |
|---|---|
| author | Zhiguo Xin |
| author_facet | Zhiguo Xin |
| author_sort | Zhiguo Xin |
| collection | DOAJ |
| description | Table tennis is a highly technical sport. Beginners often struggle to correct and optimize their movements because they lack professional knowledge and guidance. To enhance the skills of beginners, a diagnosis model for table tennis beginner movement problems based on sensors and an improved dynamic time-warping algorithm is proposed. This model combines motion data collected by the Azure Kinect sensors. First, the data is pre-processed. Then, missing data is filled in using segmented cubic spline interpolation. Next, the mean filtering algorithm is used to denoise the data. Finally, the start and end positions of each action are determined to achieve effective action segmentation. The body pose vectors are reconstructed and normalized, and finally, the action category templates are created using Euclidean centroids. The experimental results show that the action detection accuracy of this model is as high as 99.54 %, which is superior to the comparison model. After training with the diagnostic results of the model, the similarity of beginners' motion trajectories in the Z-axis direction increases from below 92 % to 97.90 %, and the similarity of their movement speed increases from below 85 % to 97.97 %. This study diagnoses the movement problems of beginner table tennis players through technical means and improves their technical level, providing a new technical approach to table tennis training. |
| format | Article |
| id | doaj-art-35c19a9081bf4f82b31601238360a4b0 |
| institution | OA Journals |
| issn | 2772-9419 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Systems and Soft Computing |
| spelling | doaj-art-35c19a9081bf4f82b31601238360a4b02025-08-20T02:36:07ZengElsevierSystems and Soft Computing2772-94192025-12-01720035010.1016/j.sasc.2025.200350Diagnosis model for action problems of table tennis beginners based on improved DTW algorithmZhiguo Xin0Corresponding author.; Department of PE, Wuxi Vocational Institute of Arts & Technology, Yixing 214200, PR ChinaTable tennis is a highly technical sport. Beginners often struggle to correct and optimize their movements because they lack professional knowledge and guidance. To enhance the skills of beginners, a diagnosis model for table tennis beginner movement problems based on sensors and an improved dynamic time-warping algorithm is proposed. This model combines motion data collected by the Azure Kinect sensors. First, the data is pre-processed. Then, missing data is filled in using segmented cubic spline interpolation. Next, the mean filtering algorithm is used to denoise the data. Finally, the start and end positions of each action are determined to achieve effective action segmentation. The body pose vectors are reconstructed and normalized, and finally, the action category templates are created using Euclidean centroids. The experimental results show that the action detection accuracy of this model is as high as 99.54 %, which is superior to the comparison model. After training with the diagnostic results of the model, the similarity of beginners' motion trajectories in the Z-axis direction increases from below 92 % to 97.90 %, and the similarity of their movement speed increases from below 85 % to 97.97 %. This study diagnoses the movement problems of beginner table tennis players through technical means and improves their technical level, providing a new technical approach to table tennis training.http://www.sciencedirect.com/science/article/pii/S2772941925001681Table tennisSensorsAction detectionData processingProblem diagnosisSimilarity |
| spellingShingle | Zhiguo Xin Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm Systems and Soft Computing Table tennis Sensors Action detection Data processing Problem diagnosis Similarity |
| title | Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm |
| title_full | Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm |
| title_fullStr | Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm |
| title_full_unstemmed | Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm |
| title_short | Diagnosis model for action problems of table tennis beginners based on improved DTW algorithm |
| title_sort | diagnosis model for action problems of table tennis beginners based on improved dtw algorithm |
| topic | Table tennis Sensors Action detection Data processing Problem diagnosis Similarity |
| url | http://www.sciencedirect.com/science/article/pii/S2772941925001681 |
| work_keys_str_mv | AT zhiguoxin diagnosismodelforactionproblemsoftabletennisbeginnersbasedonimproveddtwalgorithm |