Optimizing Athlete Performance with Sports Activity Analysis

 




 

Tham, Ming Keat (2018) Optimizing Athlete Performance with Sports Activity Analysis. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
Tham Ming Keat_FULL TEXT.pdf
Restricted to Registered users only

Download (2MB)

Abstract

Badminton is a popular sport that is constantly growing in Malaysia. However, with the growth of badminton, there is still a lack of computer vision solutions that assists coaches and trainers in order to optimize and analyze the athlete’s performance. This is in contrast to basketball, hockey and football where there are existing computer vision solutions that can track players and their actions to provide analytics and information to the coaches and even players. These analytics can be used to study the players and their performances during a match. The project aims to build a computer vision solution to assists coaches in order to optimize badminton athlete’s performance. A convolutional neural network (CNN) with OpenPose library installed is used to detect the joints of the athletes. Then, the detected joints are passed into an Artificial Neural Network (ANN) to predict the badminton strokes performed by the athlete. The badminton strokes investigated include smash, lift, net, drive and serve. In addition, the movement of the athletes are tracked as it plays an important role when performing analysis to determine the strokes performed by the athlete. The output of the system are video files with a bounding box drawn over the athlete along with the prediction of the strokes that was used by the athlete. A graph visualizing the frequency of different badminton strokes performed by the athlete are generated. The system aims to assist coaches to help optimize badminton players’ performance by providing them with informative analytics which helps them to create individualistic training plans. More information such as player stamina, movement speed, jump height and even rally duration should be implemented in future work in order to provide more detailed analytics to the coaches.

Item Type: Final Year Project
Subjects: Science > Computer Science
Geography. Anthropology. Recreation > Recreation Leisure > Sports
Faculties: Faculty of Computing and Information Technology > Bachelor of Computer Science (Honours) in Software Engineering
Depositing User: Library Editor
Date Deposited: 01 Apr 2019 09:07
Last Modified: 11 Apr 2022 06:40
URI: https://eprints.tarc.edu.my/id/eprint/1564