Article: Taekwondo movement quality evaluation algorithm based on OpenPose skeleton extraction optimisation Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.88 pp.30 - 50 Abstract: The study proposes to improve the OpenPose estimation algorithm's skeleton extraction algorithm, which is based on microgrid and merged convolution, to address the issue that the current methods for assessing the quality of taekwondo movement are not accurate or valid enough to meet practical needs. Based on this improved algorithm, an algorithm is constructed to assess the quality of taekwondo movement. Experimental results show that the optimised OpenPose algorithm achieves 92.8% accuracy, 95.2% recall, and a validation loss value of 0.72, demonstrating better overall performance than the comparison methods. The research's suggested OpenPose skeleton extraction algorithm and action quality evaluation algorithm outperform the comparison algorithm in regard to accuracy, performance, and running speed, and have useful applications. Inderscience Publishers - linking academia, business and industry through research

Title: Taekwondo movement quality evaluation algorithm based on OpenPose skeleton extraction optimisation

Authors: Jianpeng Hou; Jiakun Wu

Addresses: Heilongjiang University of Chinese Medicine, Harbin, 150040, China ' Heilongjiang University of Chinese Medicine, Harbin, 150040, China

Abstract: The study proposes to improve the OpenPose estimation algorithm's skeleton extraction algorithm, which is based on microgrid and merged convolution, to address the issue that the current methods for assessing the quality of taekwondo movement are not accurate or valid enough to meet practical needs. Based on this improved algorithm, an algorithm is constructed to assess the quality of taekwondo movement. Experimental results show that the optimised OpenPose algorithm achieves 92.8% accuracy, 95.2% recall, and a validation loss value of 0.72, demonstrating better overall performance than the comparison methods. The research's suggested OpenPose skeleton extraction algorithm and action quality evaluation algorithm outperform the comparison algorithm in regard to accuracy, performance, and running speed, and have useful applications.

Keywords: OpenPose; skeleton extraction; taekwondo; sport quality assessment; intelligence.

DOI: 10.1504/IJICT.2026.155567

International Journal of Information and Communication Technology, 2026 Vol.27 No.88, pp.30 - 50

Received: 19 Mar 2026
Accepted: 19 May 2026
Published online: 05 Aug 2026 *

Title: Taekwondo movement quality evaluation algorithm based on OpenPose skeleton extraction optimisation

Keep up-to-date