Automatic identification of subtechniques in skating-style roller skiing using inertial sensors

Yoshihisa Sakurai, Zenya Fujita, Yusuke Ishige

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

This study aims to develop and validate an automated system for identifying skating-style cross-country subtechniques using inertial sensors. In the first experiment, the performance of a male cross-country skier was used to develop an automated identification system. In the second, eight male and seven female college cross-country skiers participated to validate the developed identification system. Each subject wore inertial sensors on both wrists and both roller skis, and a small video camera on a backpack. All subjects skied through a 3450 m roller ski course using a skating style at their maximum speed. The adopted subtechniques were identified by the automated method based on the data obtained from the sensors, as well as by visual observations from a video recording of the same ski run. The system correctly identified 6418 subtechniques from a total of 6768 cycles, which indicates an accuracy of 94.8%. The precisions of the automatic system for identifying the V1R, V1L, V2R, V2L, V2AR, and V2AL subtechniques were 87.6%, 87.0%, 97.5%, 97.8%, 92.1%, and 92.0%, respectively. Most incorrect identification cases occurred during a subtechnique identification that included a transition and turn event. Identification accuracy can be improved by separately identifying transition and turn events. This system could be used to evaluate each skier’s subtechniques in course conditions.

Original languageEnglish
Article number473
JournalSensors (Switzerland)
Volume16
Issue number4
DOIs
Publication statusPublished - 2016 Apr 2

Fingerprint

skis
Skating
Skiing
rollers
system identification
Video Recording
sensors
Identification (control systems)
Sensors
Wrist
wrist
Video recording
visual observation
Video cameras
recording
cameras
Wear of materials
cycles
Experiments

Keywords

  • Accelerometer
  • Cross-country skiing
  • Gyroscope

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Atomic and Molecular Physics, and Optics
  • Analytical Chemistry
  • Biochemistry

Cite this

Automatic identification of subtechniques in skating-style roller skiing using inertial sensors. / Sakurai, Yoshihisa; Fujita, Zenya; Ishige, Yusuke.

In: Sensors (Switzerland), Vol. 16, No. 4, 473, 02.04.2016.

Research output: Contribution to journalArticle

@article{9cfbc514adf045699952486b5a416d6f,
title = "Automatic identification of subtechniques in skating-style roller skiing using inertial sensors",
abstract = "This study aims to develop and validate an automated system for identifying skating-style cross-country subtechniques using inertial sensors. In the first experiment, the performance of a male cross-country skier was used to develop an automated identification system. In the second, eight male and seven female college cross-country skiers participated to validate the developed identification system. Each subject wore inertial sensors on both wrists and both roller skis, and a small video camera on a backpack. All subjects skied through a 3450 m roller ski course using a skating style at their maximum speed. The adopted subtechniques were identified by the automated method based on the data obtained from the sensors, as well as by visual observations from a video recording of the same ski run. The system correctly identified 6418 subtechniques from a total of 6768 cycles, which indicates an accuracy of 94.8{\%}. The precisions of the automatic system for identifying the V1R, V1L, V2R, V2L, V2AR, and V2AL subtechniques were 87.6{\%}, 87.0{\%}, 97.5{\%}, 97.8{\%}, 92.1{\%}, and 92.0{\%}, respectively. Most incorrect identification cases occurred during a subtechnique identification that included a transition and turn event. Identification accuracy can be improved by separately identifying transition and turn events. This system could be used to evaluate each skier’s subtechniques in course conditions.",
keywords = "Accelerometer, Cross-country skiing, Gyroscope",
author = "Yoshihisa Sakurai and Zenya Fujita and Yusuke Ishige",
year = "2016",
month = "4",
day = "2",
doi = "10.3390/s16040473",
language = "English",
volume = "16",
journal = "Sensors (Switzerland)",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "4",

}

TY - JOUR

T1 - Automatic identification of subtechniques in skating-style roller skiing using inertial sensors

AU - Sakurai, Yoshihisa

AU - Fujita, Zenya

AU - Ishige, Yusuke

PY - 2016/4/2

Y1 - 2016/4/2

N2 - This study aims to develop and validate an automated system for identifying skating-style cross-country subtechniques using inertial sensors. In the first experiment, the performance of a male cross-country skier was used to develop an automated identification system. In the second, eight male and seven female college cross-country skiers participated to validate the developed identification system. Each subject wore inertial sensors on both wrists and both roller skis, and a small video camera on a backpack. All subjects skied through a 3450 m roller ski course using a skating style at their maximum speed. The adopted subtechniques were identified by the automated method based on the data obtained from the sensors, as well as by visual observations from a video recording of the same ski run. The system correctly identified 6418 subtechniques from a total of 6768 cycles, which indicates an accuracy of 94.8%. The precisions of the automatic system for identifying the V1R, V1L, V2R, V2L, V2AR, and V2AL subtechniques were 87.6%, 87.0%, 97.5%, 97.8%, 92.1%, and 92.0%, respectively. Most incorrect identification cases occurred during a subtechnique identification that included a transition and turn event. Identification accuracy can be improved by separately identifying transition and turn events. This system could be used to evaluate each skier’s subtechniques in course conditions.

AB - This study aims to develop and validate an automated system for identifying skating-style cross-country subtechniques using inertial sensors. In the first experiment, the performance of a male cross-country skier was used to develop an automated identification system. In the second, eight male and seven female college cross-country skiers participated to validate the developed identification system. Each subject wore inertial sensors on both wrists and both roller skis, and a small video camera on a backpack. All subjects skied through a 3450 m roller ski course using a skating style at their maximum speed. The adopted subtechniques were identified by the automated method based on the data obtained from the sensors, as well as by visual observations from a video recording of the same ski run. The system correctly identified 6418 subtechniques from a total of 6768 cycles, which indicates an accuracy of 94.8%. The precisions of the automatic system for identifying the V1R, V1L, V2R, V2L, V2AR, and V2AL subtechniques were 87.6%, 87.0%, 97.5%, 97.8%, 92.1%, and 92.0%, respectively. Most incorrect identification cases occurred during a subtechnique identification that included a transition and turn event. Identification accuracy can be improved by separately identifying transition and turn events. This system could be used to evaluate each skier’s subtechniques in course conditions.

KW - Accelerometer

KW - Cross-country skiing

KW - Gyroscope

UR - http://www.scopus.com/inward/record.url?scp=84962428155&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962428155&partnerID=8YFLogxK

U2 - 10.3390/s16040473

DO - 10.3390/s16040473

M3 - Article

VL - 16

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-3210

IS - 4

M1 - 473

ER -