Several types of sensors have been available in off-the-shelf mobile devices, including motion, magnetic, vision, acoustic, and location sensors. This paper focuses on the fusion of the data acquired from motion and magnetic sensors, i.e., accelerometer, gyroscope and magnetometer sensors, for the recognition of Activities of Daily Living (ADL). Based on pattern recognition techniques, the system developed in this study includes data acquisition, data processing, data fusion, and classification methods like Artificial Neural Networks (ANN). Multiple settings of the ANN were implemented and evaluated in which the best accuracy obtained, with Deep Neural Networks (DNN), was 89.51%. This novel approach applies L2 regularization and normalization techniques on the sensors’ data proved it suitability and reliability for the ADL recognition.
Identification of activities of daily living through data fusion on motion and magnetic sensors embedded on mobile devices / Pires, Ivan Miguel; Garcia, Nuno M.; Pombo, Nuno; Flórez-Revuelta, Francisco; Spinsante, Susanna; Teixeira, Maria Canavarro. - In: PERVASIVE AND MOBILE COMPUTING. - ISSN 1574-1192. - ELETTRONICO. - 47:(2018), pp. 78-93. [10.1016/j.pmcj.2018.05.005]
Identification of activities of daily living through data fusion on motion and magnetic sensors embedded on mobile devices
Garcia, Nuno M.Supervision
;Spinsante, SusannaWriting – Review & Editing
;
2018-01-01
Abstract
Several types of sensors have been available in off-the-shelf mobile devices, including motion, magnetic, vision, acoustic, and location sensors. This paper focuses on the fusion of the data acquired from motion and magnetic sensors, i.e., accelerometer, gyroscope and magnetometer sensors, for the recognition of Activities of Daily Living (ADL). Based on pattern recognition techniques, the system developed in this study includes data acquisition, data processing, data fusion, and classification methods like Artificial Neural Networks (ANN). Multiple settings of the ANN were implemented and evaluated in which the best accuracy obtained, with Deep Neural Networks (DNN), was 89.51%. This novel approach applies L2 regularization and normalization techniques on the sensors’ data proved it suitability and reliability for the ADL recognition.File | Dimensione | Formato | |
---|---|---|---|
2018_Pires_etal_PervasiveMobileComp_accepted.pdf
accesso aperto
Descrizione: Versione post print
Tipologia:
Documento in post-print (versione successiva alla peer review e accettata per la pubblicazione)
Licenza d'uso:
Creative commons
Dimensione
614.41 kB
Formato
Adobe PDF
|
614.41 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.