Footwork recognition and trajectory tracking in track and field based on image processing
Abstract In track and field sports, footwork can greatly affect the effect and performance of sports. Accurate footwork can effectively improve the performance of professional athletes, and for ordinary trainers, it can reduce the probability of training injuries. To solve the problem that tradition...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-03-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-95570-1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850208563925876736 |
|---|---|
| author | Jiaju Zhu Zhong Zhang Runnan Liu Junyi Liu |
| author_facet | Jiaju Zhu Zhong Zhang Runnan Liu Junyi Liu |
| author_sort | Jiaju Zhu |
| collection | DOAJ |
| description | Abstract In track and field sports, footwork can greatly affect the effect and performance of sports. Accurate footwork can effectively improve the performance of professional athletes, and for ordinary trainers, it can reduce the probability of training injuries. To solve the problem that traditional footwork is inaccurate and not well accepted by people, this paper has used an image processing method based on support vector machine (SVM) algorithm to identify and track the footwork. In this paper, a 13-s video image was extracted frame by frame from the athletes’ videos in Olympic sports competitions, and the athletes’ footwork was used as a benchmark to track their motion trajectories, extracting the corresponding feature points and categorizing them. 10 school athletes, 6 males and 4 females, were selected to track their movement pace and trajectory with a camera. The behaviors were standardized according to the extracted features, and the behaviors before and after standardization were compared. The results showed that the SVM algorithm had the most stable classification accuracy, higher recognition accuracy and better performance compared with other classification algorithms. Image processing of standardized track and field movements was effective in improving athletes’ performance, with all 10 athletes tested improving their performance between 0.4 and 0.6. The SVM algorithm-based image processing method is more acceptable after validation of its effectiveness, and the method can be extended more easily. |
| format | Article |
| id | doaj-art-8a36a75d648b4d728111c895d655241c |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-8a36a75d648b4d728111c895d655241c2025-08-20T02:10:13ZengNature PortfolioScientific Reports2045-23222025-03-0115111110.1038/s41598-025-95570-1Footwork recognition and trajectory tracking in track and field based on image processingJiaju Zhu0Zhong Zhang1Runnan Liu2Junyi Liu3School of Physical Education, Northeast Normal UniversitySchool of Mechanical and Aerospace Engineering, Jilin UniversitySchool of Ship Engineering, Harbin Engineering UniversitySchool of Physical Education, Northeast Normal UniversityAbstract In track and field sports, footwork can greatly affect the effect and performance of sports. Accurate footwork can effectively improve the performance of professional athletes, and for ordinary trainers, it can reduce the probability of training injuries. To solve the problem that traditional footwork is inaccurate and not well accepted by people, this paper has used an image processing method based on support vector machine (SVM) algorithm to identify and track the footwork. In this paper, a 13-s video image was extracted frame by frame from the athletes’ videos in Olympic sports competitions, and the athletes’ footwork was used as a benchmark to track their motion trajectories, extracting the corresponding feature points and categorizing them. 10 school athletes, 6 males and 4 females, were selected to track their movement pace and trajectory with a camera. The behaviors were standardized according to the extracted features, and the behaviors before and after standardization were compared. The results showed that the SVM algorithm had the most stable classification accuracy, higher recognition accuracy and better performance compared with other classification algorithms. Image processing of standardized track and field movements was effective in improving athletes’ performance, with all 10 athletes tested improving their performance between 0.4 and 0.6. The SVM algorithm-based image processing method is more acceptable after validation of its effectiveness, and the method can be extended more easily.https://doi.org/10.1038/s41598-025-95570-1Track and field sportsFootwork recognitionSIFT (scale-invariant feature transform) feature extractionSVM algorithm |
| spellingShingle | Jiaju Zhu Zhong Zhang Runnan Liu Junyi Liu Footwork recognition and trajectory tracking in track and field based on image processing Scientific Reports Track and field sports Footwork recognition SIFT (scale-invariant feature transform) feature extraction SVM algorithm |
| title | Footwork recognition and trajectory tracking in track and field based on image processing |
| title_full | Footwork recognition and trajectory tracking in track and field based on image processing |
| title_fullStr | Footwork recognition and trajectory tracking in track and field based on image processing |
| title_full_unstemmed | Footwork recognition and trajectory tracking in track and field based on image processing |
| title_short | Footwork recognition and trajectory tracking in track and field based on image processing |
| title_sort | footwork recognition and trajectory tracking in track and field based on image processing |
| topic | Track and field sports Footwork recognition SIFT (scale-invariant feature transform) feature extraction SVM algorithm |
| url | https://doi.org/10.1038/s41598-025-95570-1 |
| work_keys_str_mv | AT jiajuzhu footworkrecognitionandtrajectorytrackingintrackandfieldbasedonimageprocessing AT zhongzhang footworkrecognitionandtrajectorytrackingintrackandfieldbasedonimageprocessing AT runnanliu footworkrecognitionandtrajectorytrackingintrackandfieldbasedonimageprocessing AT junyiliu footworkrecognitionandtrajectorytrackingintrackandfieldbasedonimageprocessing |