Classification of human activity is an increasingly popular topic, as it is employed in various fields from fitness to remote health monitoring. Current automated approaches based on wearable sensors typically use supervised learning methodologies, where a classifier is trained with experimental data. This paper proposes the use of body motion and sensor simulation for building, or extending, the training databases and improve the classifier accuracy, without requiring further experimental campaigns.

Training a classifier for activity recognition using body motion simulation / Grosso, M.; Lena, D.; Rinaudo, S.; Guzman, D. A. F.; Demarchi, D.. - 2018:(2018), pp. 1-4. (Intervento presentato al convegno 2017 IEEE Biomedical Circuits and Systems Conference, BioCAS 2017 tenutosi a Politecnico di Torino, ita nel 2017) [10.1109/BIOCAS.2017.8325117].

Training a classifier for activity recognition using body motion simulation

Grosso M.;Guzman D. A. F.;Demarchi D.
2018

Abstract

Classification of human activity is an increasingly popular topic, as it is employed in various fields from fitness to remote health monitoring. Current automated approaches based on wearable sensors typically use supervised learning methodologies, where a classifier is trained with experimental data. This paper proposes the use of body motion and sensor simulation for building, or extending, the training databases and improve the classifier accuracy, without requiring further experimental campaigns.
2018
978-1-5090-5803-7
File in questo prodotto:
File Dimensione Formato  
08325117.pdf

non disponibili

Descrizione: Guzman - BioCAS17
Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 272.02 kB
Formato Adobe PDF
272.02 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2845677