Title: Athletes' throwing action recognition method based on PCA-LBP algorithm

Authors: Yingjie Liu

Addresses: School of Physical Education, Shangqiu Normal University, Shangqiu 476000, China

Abstract: In order to improve the average recognition rate and accuracy of motion recognition, and improve the recognition effect, a method of athlete throwing motion recognition based on PCA-LBP algorithm is proposed. Kinect device is used to collect motion images, and filter the collected images to reduce the noise contained in the images; LBP method is used to collect the features of the pre processed moving images, obtain the feature codes, and PCA method is used to reduce the dimensions of the extracted motion features. The motion image is segmented to obtain the throwing action area of the athlete, and the action recognition is conducted based on the segmentation results to obtain the throwing action recognition results of the athlete. The analysis of experimental results shows that the method in this paper effectively improves the average recognition rate and recognition accuracy of actions, and the recognition effect is good.

Keywords: PCA-LBP algorithm; action recognition; Kinect device; feature dimension reduction; PCA method.

DOI: 10.1504/IJCISTUDIES.2023.132493

International Journal of Computational Intelligence Studies, 2023 Vol.12 No.1/2, pp.130 - 141

Received: 18 Nov 2022
Accepted: 21 Jan 2023
Published online: 24 Jul 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article