A multifactorial detection model of young children’s physical abnormality based on image recognition technology under the concept of physical and health integration
Author:
Chen Jundan1, Zhang Lifang2
Affiliation:
1. School of Physical Education, Hunan University of Arts and Science , Changde , Hunan, , China . 2. School of Physical Education Science, Changsha Normal University , Changsha , Hunan, , China .
Abstract
Abstract
This paper analyzes the influence of “the concept of integration of body and health” on the physical development of young children and proposes that scientific sports are an important way to intervene in the poor physical appearance of young children. In order to facilitate and accurately identify young children with poor posture, the boundary tracking algorithm is used to detect the human body contour, and based on the physical characteristics, a human posture recognition algorithm based on multi-feature fusion and image similarity is proposed. To model common toddler body postures and obtain toddler posture features, a star model is employed. Combined with the advantage of the SVM classifier, the principal component analysis algorithm is used to design a classification system for abnormal behaviors of toddlers, which combines the static and dynamic posture data images of toddlers to detect abnormalities of toddlers’ body postures. When the difference between the angle of the left elbow to the left shoulder and the angle of the right elbow to the right shoulder to the left shoulder is about 16.58°, the toddler has a high-low shoulder posture. The detection accuracy of the method in this paper meets the needs of young children’s posture detection, and the misjudgment rate reaches 10.25%, which can assist in detecting the abnormalities of young children’s posture and facilitate the teachers and staff to carry out the sports related to young children’s posture.
Publisher
Walter de Gruyter GmbH
Reference20 articles.
1. Brock, A., De, S., Smith, S. L., & Simonyan, K. (2021, July). High-performance large-scale image recognition without normalization. In International conference on machine learning (pp. 1059-1071). PMLR. 2. Zoph, B., Vasudevan, V., Shlens, J., & Le, Q. V. (2018). Learning transferable architectures for scalable image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 8697-8710). 3. McCay, K. D., Ho, E. S., Shum, H. P., Fehringer, G., Marcroft, C., & Embleton, N. D. (2020). Abnormal infant movements classification with deep learning on pose-based features. IEEE Access, 8, 51582-51592. 4. Hesse, N., Pujades, S., Romero, J., Black, M. J., Bodensteiner, C., Arens, M., ... & Sebastian Schroeder, A. (2018). Learning an infant body model from RGB-D data for accurate full body motion analysis. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part I (pp. 792-800). Springer International Publishing. 5. Hesse, N., Pujades, S., Black, M. J., Arens, M., Hofmann, U. G., & Schroeder, A. S. (2019). Learning and tracking the 3D body shape of freely moving infants from RGB-D sequences. IEEE transactions on pattern analysis and machine intelligence, 42(10), 2540-2551.
|
|