RESEARCH ARTICLE


Kinematic Analysis of the Reconstruction Error of A Calibration Volume for 3d Analysis in Swimming



S. Brandao, P. Figueiredo, P. Goncalves, J. P. Vilas-Boas, R. J. Fernandes*
University of Porto, Faculty of Sport, Cifi2d, Porto, Portugal


Article Metrics

CrossRef Citations:
0
Total Statistics:

Full-Text HTML Views: 415
Abstract HTML Views: 647
PDF Downloads: 395
Total Views/Downloads: 1457
Unique Statistics:

Full-Text HTML Views: 223
Abstract HTML Views: 356
PDF Downloads: 247
Total Views/Downloads: 826



Creative Commons License
© 2010 Brandao et al.

open-access license: This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0), a copy of which is available at: https://creativecommons.org/licenses/by/4.0/legalcode. This license permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

* Address correspondence to this author at the Faculty of Sport, Porto University, Portugal; Tel: +351 225074763; E-mail: Ricfer@fade.up.pt


Abstract

The purpose of this study was to assess the accuracy and reliability of the three-dimensional reconstruction - above and below water - of a calibration volume for three-dimensional analysis in swimming. The calibration volume (3x3x3m) was placed in a 25m x 12.5m x 2m swimming pool and was recorded simultaneously by 4 under and 2 above water synchronized cameras. To assess the number of control points required to maximise the accuracy of threedimensional coordinate reconstruction, 12 markers in the calibrated space were digitised over 10 fields for each underwater and above water camera views. Seven series of 12 markers, using 8, 12, 16, 20, 24, 28 and 30 control points were digitised. Direct linear transformation methods were used to estimate the marker locations on the volume. Comparison among different numbers of control points showed that the set of 20 (underwater) and 16 (above water) points produced the most accurate results. The average root mean square errors were (x, y and z, respectively): (i) 4.85mm, 2.52mm and 7.43mm (set of 20 digitised underwater points) and (ii) 4.11mm, 5.69mm and 3.90mm (set of 16 digitised above water points). The standard deviation in underwater cameras was 1.22mm, 0.33mm and 3.47mm, and 1.57mm, 2.63mm and 2.35mm for above water cameras (for x, y and z, respectively). The calibration volume was found to have high accuracy and reliability.

Keywords: Above water, accuracy, biomechanics, reliability, three-dimensional, underwater.