RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY
Supervised learning as a sub-discipline of machine learning enables the recognition of correlations between input variables (features) and associated outputs (classes) and the application of these to previously unknown data sets. In addition to typical areas of application such as speech and image r...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Polish Association for Knowledge Promotion
2023-03-01
|
| Series: | Applied Computer Science |
| Subjects: | |
| Online Access: | http://www.acs.pollub.pl/pdf/v19n1/10.pdf |
| _version_ | 1797848878704230400 |
|---|---|
| author | Pascal KRUTZ Matthias REHM Holger SCHLEGEL Martin DIX |
| author_facet | Pascal KRUTZ Matthias REHM Holger SCHLEGEL Martin DIX |
| author_sort | Pascal KRUTZ |
| collection | DOAJ |
| description | Supervised learning as a sub-discipline of machine learning enables the recognition of correlations between input variables (features) and associated outputs (classes) and the application of these to previously unknown data sets. In addition to typical areas of application such as speech and image recognition, fields of applications are also being developed in the sports and fitness sector. The purpose of this work is to implement a workflow for the automated recognition of sports exercises in the Matlab® program¬ming environment and to carry out a comparison of different model structures. First, the acquisition of the sensor signals provided in the local network and their processing is implemented. Realised functionalities include the interpolation of lossy time series, the labelling of the activity intervals performed and, in part, the generation of sliding windows with statistical parameters. The preprocessed data are used for the training of classifiers and artificial neural networks (ANN). These are iteratively optimised in their corresponding hyper parameters for the data structure to be learned. The most reliable models are finally trained with an increased data set, validated and compared with regard to the achieved performance. In addition to the usual evaluation metrics such as F1 score and accuracy, the temporal behaviour of the assignments is also displayed graphically, allowing statements to be made about potential causes of incorrect assignments. In this context, especially the transition areas between the classes are detected as erroneous assignments as well as exercises with insufficient or clearly deviating execution. The best overall accuracy achieved with ANN and the increased dataset was 93.7 %. |
| first_indexed | 2024-04-09T18:34:46Z |
| format | Article |
| id | doaj.art-eeddba7bd5fb45029a4da5d3f8d5dc2d |
| institution | Directory Open Access Journal |
| issn | 1895-3735 2353-6977 |
| language | English |
| last_indexed | 2024-04-09T18:34:46Z |
| publishDate | 2023-03-01 |
| publisher | Polish Association for Knowledge Promotion |
| record_format | Article |
| series | Applied Computer Science |
| spelling | doaj.art-eeddba7bd5fb45029a4da5d3f8d5dc2d2023-04-11T12:14:26ZengPolish Association for Knowledge PromotionApplied Computer Science1895-37352353-69772023-03-0119115216310.35784/acs-2023-10RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGYPascal KRUTZ0Matthias REHM1Holger SCHLEGEL2Martin DIX 3Professorship for Production Systems and Processes, Chemnitz University of Technology, Chemnitz, GermanyProfessorship for Production Systems and Processes, Chemnitz University of Technology, Chemnitz, GermanyProfessorship for Production Systems and Processes, Chemnitz University of Technology, Chemnitz, GermanyFraunhofer Institute for Machine Tools and Forming Technology IWU, Chemnitz, GermanySupervised learning as a sub-discipline of machine learning enables the recognition of correlations between input variables (features) and associated outputs (classes) and the application of these to previously unknown data sets. In addition to typical areas of application such as speech and image recognition, fields of applications are also being developed in the sports and fitness sector. The purpose of this work is to implement a workflow for the automated recognition of sports exercises in the Matlab® program¬ming environment and to carry out a comparison of different model structures. First, the acquisition of the sensor signals provided in the local network and their processing is implemented. Realised functionalities include the interpolation of lossy time series, the labelling of the activity intervals performed and, in part, the generation of sliding windows with statistical parameters. The preprocessed data are used for the training of classifiers and artificial neural networks (ANN). These are iteratively optimised in their corresponding hyper parameters for the data structure to be learned. The most reliable models are finally trained with an increased data set, validated and compared with regard to the achieved performance. In addition to the usual evaluation metrics such as F1 score and accuracy, the temporal behaviour of the assignments is also displayed graphically, allowing statements to be made about potential causes of incorrect assignments. In this context, especially the transition areas between the classes are detected as erroneous assignments as well as exercises with insufficient or clearly deviating execution. The best overall accuracy achieved with ANN and the increased dataset was 93.7 %.http://www.acs.pollub.pl/pdf/v19n1/10.pdfhuman activity recognitionmachine learningneural networksclassifier |
| spellingShingle | Pascal KRUTZ Matthias REHM Holger SCHLEGEL Martin DIX RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY Applied Computer Science human activity recognition machine learning neural networks classifier |
| title | RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY |
| title_full | RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY |
| title_fullStr | RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY |
| title_full_unstemmed | RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY |
| title_short | RECOGNITION OF SPORTS EXERCISES USING INERTIAL SENSOR TECHNOLOGY |
| title_sort | recognition of sports exercises using inertial sensor technology |
| topic | human activity recognition machine learning neural networks classifier |
| url | http://www.acs.pollub.pl/pdf/v19n1/10.pdf |
| work_keys_str_mv | AT pascalkrutz recognitionofsportsexercisesusinginertialsensortechnology AT matthiasrehm recognitionofsportsexercisesusinginertialsensortechnology AT holgerschlegel recognitionofsportsexercisesusinginertialsensortechnology AT martindix recognitionofsportsexercisesusinginertialsensortechnology |