Automatic detection of pronation and supination of runners using computer vision

Anže Pečar (2014) Automatic detection of pronation and supination of runners using computer vision. EngD thesis.

Preview

Abstract

Running is a popular form of recreation. Running clubs and shops commonly provide services related to video analysis of running. The most usual service is determining pronation type of a runner. The normal method is measuring the eversion angle of an ankle, usually measured manually from a backside video of a runner on a treadmill. Therefore, there is a need to develop applications for automatic analysis of running gait. In this thesis we have developed methods for automatic measurement of the eversion angle of an ankle of runners on a treadmill, which automatically measure the angle and the type of pronation. Tested on a group of 15 runners, these methods produced reasonably accurate result for roughly determining the pronation type. However they are not accurate enough for wider commercial use and for research in sports. We propose some guidelines and improvements for further work to minimise the shortcomings and enable efficient use in applications for general use and in research.

Item Type:

Thesis (EngD thesis)

Keywords:

gait analysis, running, motion detection, human skeleton, automated video analysis, computer vision

Number of Pages:

Language of Content:

Slovenian

Mentor / Comentors:

Name and Surname	ID	Function
doc. dr. Peter Peer	294	Mentor

Link to COBISS:

http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=00010687060)

Institution:

University of Ljubljana

Department:

Faculty of Computer and Information Science

Item ID:

2597

Date Deposited:

02 Jul 2014 15:21

Last Modified:

28 Jul 2014 09:54

URI:

http://eprints.fri.uni-lj.si/id/eprint/2597

Actions (login required)

View Item