SEGMENTATION OF HUMAN MOTION ACCELERATION WITH PROBABILISTIC CLASSIFIER
Abstract eng: This paper describes a method for signal segmentation in human motion analysis. Proposed method uses a probabilistic change point estimator combined with a Trigg’s tracking signal for detection of changes in a signal variation and segmentation to the subsections by these change points. Main usage of this method is in fields of sport training or health condition monitoring but it can be also used in technical monitoring.
Publisher:
Brno University of Technology, Institute of Solid Mechanics, Mechatronics and Biomechanics, Brno
Conference Title:
Conference Title:
Engineering Mechanics 2017
Conference Venue:
Svratka, CZ
Conference Dates:
2017-05-15 / 2017-05-18
Rights:
Text je chráněný podle autorského zákona č. 121/2000 Sb.
Record appears in:
Record created 2017-05-22, last modified 2017-05-22