The Open UniversitySkip to content
 

A Novel Multi-View Table Tennis Umpiring Framework

Myint, Hnin (2019). A Novel Multi-View Table Tennis Umpiring Framework. PhD thesis The Open University.

Full text available as:
[img]
Preview
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (12MB) | Preview
Google Scholar: Look up in Google Scholar

Abstract

This research investigates the development of a low-cost multi-view umpiring framework, as an alternative to the current expensive systems that are almost exclusively restricted to elite professional sports. Table tennis has been selected as the testbed because, while automating the process is challenging, it has many different complex match elements including the service, return and rallies, which are governed by a strict set of regulations. The focus is mainly on the rally element rather than the whole match. Ball detection and tracking in video frames are undertaken to determine reliably the ball position relative to key reference objects like the table surface and net, and the ball’s flight path is used to determine the rally’s status.

While a low-cost option has benefits, it is technically challenging due to the limited number of cameras and generally low video resolution used. This thesis presents a portable multi-view umpiring framework that identifies each state change in a rally. It makes three significant contributions to knowledge: i) a reliable ball detection strategy that accurately detects the location of the ball in low-resolution sequences; ii) a novel framework for ball tracking using a multi-view system, and iii) a new state-machine based evaluation system for analysing table tennis rallies.

In a series of ten different test scenarios, the system achieved an average of 94% system detection rate and 100% accurate decisions. A test sequence of duration 1 s can be processed in 8 s, leading to a delay of only 7 s, which is considered acceptable for practical purposes. This solution has the potential to reform the way matches are umpired, providing objectivity in resolving disputed decisions. It affords an economic technology for amateur players, while the multi-view facility is extendible to other relevant ball-based sports. Finally, the ball flight path analysis mechanism can be a valuable training tool for skills development.

Item Type: Thesis (PhD)
Copyright Holders: 2018 The Author
Keywords: computer vision; artificial intelligence; expert systems; table tennis umpiring; multi-agents; agent systems; object detection and tracking; image processing
Academic Unit/School: Faculty of Science, Technology, Engineering and Mathematics (STEM)
Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
Related URLs:
Item ID: 60470
Depositing User: Hnin Myint
Date Deposited: 13 May 2019 14:58
Last Modified: 11 Jul 2019 03:23
URI: http://oro.open.ac.uk/id/eprint/60470
Share this page:

Download history for this item

These details should be considered as only a guide to the number of downloads performed manually. Algorithmic methods have been applied in an attempt to remove automated downloads from the displayed statistics but no guarantee can be made as to the accuracy of the figures.

Actions (login may be required)

Policies | Disclaimer

© The Open University   contact the OU