Title: Image error correction of hockey players' step-by-step pull shooting based on Bayesian classification
Authors: Hongping Li
Addresses: College of Sports Science, Shenyang Normal University, Shenyang, 110034, Liaoning, China
Abstract: The target recognition accuracy of the traditional motion image error correction method is low, which leads to its poor application effect. In order to solve this problem, this paper proposes a new image error correction method based on Bayesian classification for hockey players' step-by-step pull shooting action. The SLIC method is used to distinguish the hockey players' step-by-step pull shooting action from the image background area, to obtain the proportion of the hockey players' step-by-step pull shooting action in the super pixel area. The hockey players' step-by-step pull shooting action with incomplete Naive Bayesian classification model, and to correct the image error with interpolation method. The experimental results show that the accuracy of the method for the hockey players' step-by-step pull shooting action is higher than 98%, and the image quality is high after the error correction.
Keywords: Bayes classification; hockey players; step-by-step pull shooting; image error correction; SLIC method; interpolation method.
DOI: 10.1504/IJCSM.2021.118797
International Journal of Computing Science and Mathematics, 2021 Vol.14 No.2, pp.185 - 195
Received: 06 Jan 2020
Accepted: 02 Mar 2020
Published online: 08 Nov 2021 *