Gaussian Mixture Model Based Player Tracking Technique in Basketball Sports Video

被引：0

作者：

Jia, Xin-Hui ^{[1
]}

Evans, Cawlton ^{[2
]}

机构：

[1] Department of Art and Sports, Huanghe Science and Technology University, Zhengzhou,450061, China

[2] Faculty of Engineering, Dalhousie University, Halifax,B3H 4R2, Canada

来源：

Journal of Network Intelligence | 2024年 / 9卷 / 02期

关键词：

As sporting events become more and more competitive; many coaches are beginning to use information technology to improve the professionalism of their athletes. Monitoring and recognition of moving objects in sports video is currently a more common method. However; the uniqueness of basketball sports video poses a great challenge to target tracking techniques; especially the interference of camera shake and noise. This works suggests an athlete tracking method based on an improved Gaussian mixture model as a solution to the concerns mentioned above. Firstly; the problem of stereo visual perception in basketball sports videos is investigated. Secondly; the background modelling method based on the traditional Gaussian mixture model is analysed; and the background model parameter estimation method is constructed using the Student-t distribution. The parameter estimation is completed by the expectation-maximum algorithm and the parameter space is partitioned. Then; in order to further improve the accuracy of target tracking; a particle filtering optimization algorithm based on genetic algorithm was proposed in order to eliminate the particle degradation. Finally; two video sequences of NBA regular season games were used for target tracking tests. The experimental results show that the proposed improved Gaussian mixture model has better target tracking results compared with the traditional Gaussian mixture model and the generalised Gaussian mixture model. The tracking accuracy of the athletes is higher; which validates the effectiveness and advancedness of the proposed model. © 2024 ISSN 2414-8105;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

引用

页码：1210 / 1227

共 50 条

[21] A new compressive sensing video coding framework based on Gaussian mixture model
Li, Xiangwei
Lan, Xuguang
Yang, Meng
Xue, Jianru
Zheng, Nanning
Signal Processing: Image Communication, 2017, 55 : 66 - 79
[22] Gaussian mixture model based phase prior learning for video motion estimation
Cai, Enjian
Zhang, Yi
Mechanical Systems and Signal Processing, 2022, 175
[23] Video Smoke Detection Based on Gaussian Mixture Model and Convolutional Neural Network
Li Peng
Zhang Yan
LASER & OPTOELECTRONICS PROGRESS, 2019, 56 (21)
[24] Gaussian mixture model based phase prior learning for video motion estimation
Cai, Enjian
Zhang, Yi
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2022, 175
[25] Real Time Colour Based Player Tracking in Indoor Sports
Santiago, Catarina B.
Sousa, Armando
Reis, Luis Paulo
Estriga, Maria Luisa
COMPUTATIONAL VISION AND MEDICAL IMAGE PROCESSING: RECENT TRENDS, 2011, 19 : 17 - 35
[26] Crawling robot manipulator tracking based on gaussian mixture model of machine vision
Lou J.
Neural Computing and Applications, 2022, 34 (9) : 6683 - 6693
[27] Automatic Vehicle Detection and Motion Path Tracking Based on Gaussian Mixture Model
Roy, Kyamelia
Saha, Soham
Mondal, Tanmoy
Choudhury, Sheli Sinha
ADVANCES IN OPTICAL SCIENCE AND ENGINEERING, 2017, 194 : 669 - 679
[28] Improved Gaussian mixture model in video motion detection
Xie, Yong
Journal of Multimedia, 2013, 8 (05): : 527 - 533
[29] Improved Gaussian Mixture Model for the Task of Object Tracking
Sicre, Ronan
Nicolas, Henri
COMPUTER ANALYSIS OF IMAGES AND PATTERNS: 14TH INTERNATIONAL CONFERENCE, CAIP 2011, PT 2, 2011, 6855 : 389 - 396
[30] Gaussian Mixture Model on Tensor Field for Visual Tracking
Zhan, Xueliang
Ma, Bo
IEEE SIGNAL PROCESSING LETTERS, 2012, 19 (11) : 733 - 736

← 1 2 3 4 5 →