Skip to Main content Skip to Navigation
Reports

"Etude algorithmique préliminaire" à la conception du module eMouve au sein d’ActivCollector : détection des activités physiques en conditions habituelles de vie à l’aide de smartphones

Abstract : Obesity and sedentary lifestyles are constantly increasing for many years. INRA currently uses expensive and/or complex methods in order to estimate energy expenditure in controlled conditions in laboratory or in free-living conditions. The aim of the project is to recognize low to moderate physical activities and to estimate volunteers’energy expenditure in free-living conditions. This is possible thanks to smartphones, well-spread into the population, which are equipped of a triaxial acceleromete. The activities studied in this project are: walking, running, walking up stairs, walking down stairs, sitting, standing and driving a car. To reach this goal innovative mathematical tools have been introduced and new algorithms are proposed. The obtained recognition percentages are satisfactory compared to the results of other publications which often used several research-accelerometers unavailable to the general public. Car driving was the activity the most badly recognized (36%) and running the best recognized (97%). On average, the model recognized 79.7% of the activities made by the volunteer. Moreover, the estimation of the energy expenditure seemed to be quite close to the estimations of the reference sensors Actiheart (less than 2% of difference) and farther from SenseWear Armband estimation (about 17% of difference).The next step consists in testing the model on the data of ten volunteers and improving it.
Complete list of metadata

https://hal.inrae.fr/hal-02804185
Contributor : Migration Prodinra Connect in order to contact the contributor
Submitted on : Friday, June 5, 2020 - 11:10:43 PM
Last modification on : Friday, June 18, 2021 - 4:06:09 PM

File

2012_R_Guidoux_EtudePreliminai...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02804185, version 1
  • PRODINRA : 169524

Collections

Citation

Gérard Fleury, Romain Guidoux, Philippe Lacomme, Sylvie Rousset. "Etude algorithmique préliminaire" à la conception du module eMouve au sein d’ActivCollector : détection des activités physiques en conditions habituelles de vie à l’aide de smartphones. [Rapport Technique] RR-12-07, Université Blaise Pascal (Clermont Ferrand 2) (UBP). 2012. ⟨hal-02804185⟩

Share

Metrics

Record views

98

Files downloads

63