Abstract
The recognition technology using conventional sensors or image processing is effective for static gesture recognition, but for dynamic gesture recognition, the moving object can not be tracked in time, resulting in low recognition accuracy and efficiency. In order to optimize the above problems, the dynamic recognition method of track and field posture based on mobile monitoring technology is studied. Set up mobile monitoring equipment in the movement area and the movement track to acquire the data of track and field movement posture. After de-noising the track and field posture data, a Gaussian model is established to segment the image background. Based on the human skeleton model, the motion posture features are extracted. Using BP neural network improved by artificial fish swarm to classify the input movement posture data, the recognition of track and field movement posture is realized. The test results show that the recognition accuracy of the proposed methods is higher than 95%, the recognition efficiency is greatly improved, and it has good practical value.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Yang, Y., Sun, G., Wang, Z.: Human activity recognition based on android sensors. J. Nanchang Univ. (Natl. Sci.) 43(06), 616–620 (2019)
Lin, H.: Research on intelligent recognition of motion based on laser sensor. Laser J. 42(07), 84–89 (2021)
Liu, W.: Simulation of human body local feature points recognition based on machine learning. Comput. Simul. 38(06), 387–390+395 (2021)
Hongyu, Z., Chunfeng, Y., Song, X., et al.: Motion recognition based on weighted three-view motion history image coupled time segmentation. J. Electron. Measure. Instrum. 34(11), 194–203 (2020)
Wang, B., Li, D., Zhang, J., et al.: A recognition method of spread and grasp actions combining motion imagination and action observation. J. Xi’an Jiaotong Univ. 53(10), 151–158 (2019)
Ye, S.-T., Zhou, Y.-Z., Fan, H.-J., et al.: Joint learning of causality and spatio-temporal graph convolutional network for skeleton-based action recognition. Comput. Sci. 48(S2), 130–135 (2021)
Jia, X., Li, M.: Design of security monitoring system for mobile terminal distance education based on multimedia technology. Mod. Electron. Tech. 42(18), 77–80 (2019)
Juan, L., Ying, R.: Simulation of recognition of the vehicle driver’s head posture image feature. Comput. Simul. 35(1), 374–377 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
About this paper
Cite this paper
Lin, Q., Han, L. (2023). Dynamic Recognition Method of Track and Field Posture Based on Mobile Monitoring Technology. In: Fu, W., Yun, L. (eds) Advanced Hybrid Information Processing. ADHIP 2022. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol 469. Springer, Cham. https://doi.org/10.1007/978-3-031-28867-8_25
Download citation
DOI: https://doi.org/10.1007/978-3-031-28867-8_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-28866-1
Online ISBN: 978-3-031-28867-8
eBook Packages: Computer ScienceComputer Science (R0)