Similar gait action recognition using an inertial sensor

Trung Thanh Ngo, Yasushi Makihara, Hajime Nagahara, Yasuhiro Mukaigawa, Yasushi Yagi

Research output: Contribution to journalArticle

42 Citations (Scopus)

Abstract

This paper tackles a challenging problem of inertial sensor-based recognition for similar gait action classes (such as walking on flat ground, up/down stairs, and up/down a slope). We solve three drawbacks of existing methods in the case of gait actions: the action signal segmentation, the sensor orientation inconsistency, and the recognition of similar action classes. First, to robustly segment the walking action under drastic changes in various factors such as speed, intensity, style, and sensor orientation of different participants, we rely on the likelihood of heel strike computed employing a scale-space technique. Second, to solve the problem of 3D sensor orientation inconsistency when matching the signals captured at different sensor orientations, we correct the sensor's tilt before applying an orientation-compensative matching algorithm to solve the remaining angle. Third, to accurately classify similar actions, we incorporate the interclass relationship in the feature vector for recognition. In experiments, the proposed algorithms were positively validated with 460 participants (the largest number in the research field), and five similar gait action classes (namely walking on flat ground, up/down stairs, and up/down a slope) captured by three inertial sensors at different positions (center, left, and right) and orientations on the participant's waist.

Original languageEnglish
Pages (from-to)1289-1301
Number of pages13
JournalPattern Recognition
Volume48
Issue number4
DOIs
Publication statusPublished - Apr 1 2015

Fingerprint

Sensors
Stairs
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Ngo, T. T., Makihara, Y., Nagahara, H., Mukaigawa, Y., & Yagi, Y. (2015). Similar gait action recognition using an inertial sensor. Pattern Recognition, 48(4), 1289-1301. https://doi.org/10.1016/j.patcog.2014.10.012

Similar gait action recognition using an inertial sensor. / Ngo, Trung Thanh; Makihara, Yasushi; Nagahara, Hajime; Mukaigawa, Yasuhiro; Yagi, Yasushi.

In: Pattern Recognition, Vol. 48, No. 4, 01.04.2015, p. 1289-1301.

Research output: Contribution to journalArticle

Ngo, TT, Makihara, Y, Nagahara, H, Mukaigawa, Y & Yagi, Y 2015, 'Similar gait action recognition using an inertial sensor', Pattern Recognition, vol. 48, no. 4, pp. 1289-1301. https://doi.org/10.1016/j.patcog.2014.10.012
Ngo TT, Makihara Y, Nagahara H, Mukaigawa Y, Yagi Y. Similar gait action recognition using an inertial sensor. Pattern Recognition. 2015 Apr 1;48(4):1289-1301. https://doi.org/10.1016/j.patcog.2014.10.012
Ngo, Trung Thanh ; Makihara, Yasushi ; Nagahara, Hajime ; Mukaigawa, Yasuhiro ; Yagi, Yasushi. / Similar gait action recognition using an inertial sensor. In: Pattern Recognition. 2015 ; Vol. 48, No. 4. pp. 1289-1301.
@article{c57334a1965d44459788eb617d55e86b,
title = "Similar gait action recognition using an inertial sensor",
abstract = "This paper tackles a challenging problem of inertial sensor-based recognition for similar gait action classes (such as walking on flat ground, up/down stairs, and up/down a slope). We solve three drawbacks of existing methods in the case of gait actions: the action signal segmentation, the sensor orientation inconsistency, and the recognition of similar action classes. First, to robustly segment the walking action under drastic changes in various factors such as speed, intensity, style, and sensor orientation of different participants, we rely on the likelihood of heel strike computed employing a scale-space technique. Second, to solve the problem of 3D sensor orientation inconsistency when matching the signals captured at different sensor orientations, we correct the sensor's tilt before applying an orientation-compensative matching algorithm to solve the remaining angle. Third, to accurately classify similar actions, we incorporate the interclass relationship in the feature vector for recognition. In experiments, the proposed algorithms were positively validated with 460 participants (the largest number in the research field), and five similar gait action classes (namely walking on flat ground, up/down stairs, and up/down a slope) captured by three inertial sensors at different positions (center, left, and right) and orientations on the participant's waist.",
author = "Ngo, {Trung Thanh} and Yasushi Makihara and Hajime Nagahara and Yasuhiro Mukaigawa and Yasushi Yagi",
year = "2015",
month = "4",
day = "1",
doi = "10.1016/j.patcog.2014.10.012",
language = "English",
volume = "48",
pages = "1289--1301",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "4",

}

TY - JOUR

T1 - Similar gait action recognition using an inertial sensor

AU - Ngo, Trung Thanh

AU - Makihara, Yasushi

AU - Nagahara, Hajime

AU - Mukaigawa, Yasuhiro

AU - Yagi, Yasushi

PY - 2015/4/1

Y1 - 2015/4/1

N2 - This paper tackles a challenging problem of inertial sensor-based recognition for similar gait action classes (such as walking on flat ground, up/down stairs, and up/down a slope). We solve three drawbacks of existing methods in the case of gait actions: the action signal segmentation, the sensor orientation inconsistency, and the recognition of similar action classes. First, to robustly segment the walking action under drastic changes in various factors such as speed, intensity, style, and sensor orientation of different participants, we rely on the likelihood of heel strike computed employing a scale-space technique. Second, to solve the problem of 3D sensor orientation inconsistency when matching the signals captured at different sensor orientations, we correct the sensor's tilt before applying an orientation-compensative matching algorithm to solve the remaining angle. Third, to accurately classify similar actions, we incorporate the interclass relationship in the feature vector for recognition. In experiments, the proposed algorithms were positively validated with 460 participants (the largest number in the research field), and five similar gait action classes (namely walking on flat ground, up/down stairs, and up/down a slope) captured by three inertial sensors at different positions (center, left, and right) and orientations on the participant's waist.

AB - This paper tackles a challenging problem of inertial sensor-based recognition for similar gait action classes (such as walking on flat ground, up/down stairs, and up/down a slope). We solve three drawbacks of existing methods in the case of gait actions: the action signal segmentation, the sensor orientation inconsistency, and the recognition of similar action classes. First, to robustly segment the walking action under drastic changes in various factors such as speed, intensity, style, and sensor orientation of different participants, we rely on the likelihood of heel strike computed employing a scale-space technique. Second, to solve the problem of 3D sensor orientation inconsistency when matching the signals captured at different sensor orientations, we correct the sensor's tilt before applying an orientation-compensative matching algorithm to solve the remaining angle. Third, to accurately classify similar actions, we incorporate the interclass relationship in the feature vector for recognition. In experiments, the proposed algorithms were positively validated with 460 participants (the largest number in the research field), and five similar gait action classes (namely walking on flat ground, up/down stairs, and up/down a slope) captured by three inertial sensors at different positions (center, left, and right) and orientations on the participant's waist.

UR - http://www.scopus.com/inward/record.url?scp=84920669699&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920669699&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2014.10.012

DO - 10.1016/j.patcog.2014.10.012

M3 - Article

AN - SCOPUS:84920669699

VL - 48

SP - 1289

EP - 1301

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 4

ER -