With the rapid development of computer technology, body motion recognition technology based on computer vision came into being. In the research of sports field, the coach can help the athletes to carry out auxiliary training and find out the advantages and disadvantages of the athletes through the analysis of the athletes’ sports state, so as to guide athletes pertinently and improve their mastery of sports actions. However, the traditional motion recognition methods have low recognition accuracy and low recognition efficiency, and are difficult to be effectively used in sports training. Based on this, this paper studied the feature extraction and automatic recognition of sports actions based on video image technology, proposed the feature extraction method of sports actions based on bone key points and the K-Nearest-Neighbor (KNN) sports action recognition method combined with the improved Dynamic Time Warping (DTW) algorithm, and carried out experimental research on this method. The research showed that compared with the KNN-based sports action recognition method, the improved DTW-based KNN sports action recognition method had a 4.2% higher precision and a 2.4% higher recall. KNN sports movement recognition method based on improved DTW had good recognition effect.

Peiqi He. Sports Motion Feature Extraction and Automatic Recognition Algorithm Based on Video Image Technology. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 12: 106-117. https://doi.org/10.25236/AJCIS.2023.061212.

[1] Zhang, Xiaojun. "Application of human motion recognition utilizing deep learning and smart wearable device in sports." International Journal of System Assurance Engineering and Management 12.4 (2021): 835-843.

[2] Cust, Emily E. "Machine and deep learning for sport-specific movement recognition: a systematic review of model development and performance." Journal of sports sciences 37.5 (2019): 568-600.

[3] Yang, Yun. "Human movement monitoring and behavior recognition for intelligent sports using customizable and flexible triboelectric nanogenerator." Science China Technological Sciences 65.4 (2022): 826-836.

[4] Rahmad, Nur Azmina. "A survey of video based action recognition in sports." Indonesian Journal of Electrical Engineering and Computer Science 11.3 (2018): 987-993.

[5] Dong, Zhe, and Wang, Xiongying. "Automatic recognition of javelin athletes’ throwing angle based on recognisable statistical characteristics." International Journal of Biometrics 14.3-4 (2022): 408-420.

[6] Gurbuz, Sevgi Zubeyde, and Moeness G. Amin. "Radar-based human-motion recognition with deep learning: Promising applications for indoor monitoring." IEEE Signal Processing Magazine 36.4 (2019): 16-28.

[7] Jaouedi, Neziha, Noureddine Boujnah, and Med Salim Bouhlel. "A new hybrid deep learning model for human action recognition." Journal of King Saud University-Computer and Information Sciences 32.4 (2020): 447-453.

[8] Wang, Jucui. "Research on contour feature extraction method of multiple sports images based on nonlinear mechanics." Nonlinear Engineering 11.1 (2022): 347-354.

[9] Wang, Peng. "Deep learning-based human motion recognition for predictive context-aware human-robot collaboration." CIRP annals 67.1 (2018): 17-20.

[10] Hendry, Danica. "Development of a human activity recognition system for ballet tasks." Sports medicine-open 6.1 (2020): 1-10.

[11] Rangasamy, Keerthana. "Deep learning in sport video analysis: a review." TELKOMNIKA (Telecommunication Computing Electronics and Control) 18.4 (2020): 1926-1933.

[12] Kong, Longteng. "A joint framework for athlete tracking and action recognition in sports videos." IEEE transactions on circuits and systems for video technology 30.2 (2019): 532-548.

[13] Ullah, Mohib. "Attention-based LSTM network for action recognition in sports." Electronic Imaging 2021.6 (2021): 302.1-302.6.

[14] Li, Xinyu. "Human motion recognition with limited radar micro-Doppler signatures." IEEE Transactions on Geoscience and Remote Sensing 59.8 (2020): 6586-6599.

[15] Gao, Peijun, Zhao, Dan, and Chen, Xuanang. "Multi‐dimensional data modelling of video image action recognition and motion capture in deep learning framework." IET Image Processing 14.7 (2020): 1257-1264.