An Embedded ANN Raspberry PI for
Inertial Sensor Based Human Activity
Recognition
Achraf Jmal
1,2(B) , Rim Barioul
3 , Amel Meddeb Makhlouf
1 ,
Ahmed Fakhfakh
1,2 , and Olfa Kanoun
3
1 National School of Electronics and Telecommunications of Sfax,
University of Sfax, Sfax, Tunisia
Jmal.achraf@yahoo.com, {Amel.makhlouf,Ahmed.fakhfakh}@enetcom.usf.tn
2 Centre de Recherche en Num´ erique de Sfax, Laboratoire des Technologies des
Syst` emes Smart, LR16CRNS01, 3021 Sfax, Tunisia
3 Professorship of Measurement and Sensor Technology, Technische Universit¨ at
Chemnitz, Chemnitz, Germany
{Rim.barioul,Olfa.kanoun}@etit.tu-chemnitz.de
Abstract. Human Activity Recognition (HAR) is one of the critical
subjects of research in health and human machine interaction fields
in recent years. Algorithms such as Support Vector Machine (SVM),
K-Nearest Neighbors (K-NN), Decision Tree (DT) and many other algorithms were previously implemented to serve this common goal but most
of the traditional Machine learning proposed solutions were not satisfying in term of accuracy and real time testing process. For that, a human
activities analysis and recognition system with an embedded trained
ANN model on Raspberry PI for an online testing process is proposed in
this work. This paper includes a comparative study between the Artificial Neural Network (ANN) and the Recurrent Neural Network (RNN),
using signals produced by the accelerometer and gyroscope, embedded
within the BlueNRG-Tile sensor. After evaluate algorithms performance
in terms of accuracy and precision which reached an accuracy of 82%
for ANN and 99% for RNN, obtained ANN model was implemented in
a Raspberry PI for real-time predictions. Results show that the system
provides a real-time human activity recognition with an accuracy of 86%.
Keywords: Machine learning · Deep learning · HAR · Embedded
ANN · LSTM-RNN · Raspberry PI · Python
1 Introduction
Human activity recognition (HAR) refers to the automatic detection of various
physical activities performed by people in their daily lives [1]. Activity recognition can be achieved by exploiting information retrieved from sensors such as
Supported by German Academic Exchange Service, CRNS and ReDCAD.
c
The Author(s) 2020
M. Jmaiel et al. (Eds.): ICOST 2020, LNCS 12157, pp. 375–385, 2020.
https://doi.org/10.1007/978-3-030-51517-1_34
Inertial Sensor Based Human Activity
Recognition
Achraf Jmal
1,2(B) , Rim Barioul
3 , Amel Meddeb Makhlouf
1 ,
Ahmed Fakhfakh
1,2 , and Olfa Kanoun
3
1 National School of Electronics and Telecommunications of Sfax,
University of Sfax, Sfax, Tunisia
Jmal.achraf@yahoo.com, {Amel.makhlouf,Ahmed.fakhfakh}@enetcom.usf.tn
2 Centre de Recherche en Num´ erique de Sfax, Laboratoire des Technologies des
Syst` emes Smart, LR16CRNS01, 3021 Sfax, Tunisia
3 Professorship of Measurement and Sensor Technology, Technische Universit¨ at
Chemnitz, Chemnitz, Germany
{Rim.barioul,Olfa.kanoun}@etit.tu-chemnitz.de
Abstract. Human Activity Recognition (HAR) is one of the critical
subjects of research in health and human machine interaction fields
in recent years. Algorithms such as Support Vector Machine (SVM),
K-Nearest Neighbors (K-NN), Decision Tree (DT) and many other algorithms were previously implemented to serve this common goal but most
of the traditional Machine learning proposed solutions were not satisfying in term of accuracy and real time testing process. For that, a human
activities analysis and recognition system with an embedded trained
ANN model on Raspberry PI for an online testing process is proposed in
this work. This paper includes a comparative study between the Artificial Neural Network (ANN) and the Recurrent Neural Network (RNN),
using signals produced by the accelerometer and gyroscope, embedded
within the BlueNRG-Tile sensor. After evaluate algorithms performance
in terms of accuracy and precision which reached an accuracy of 82%
for ANN and 99% for RNN, obtained ANN model was implemented in
a Raspberry PI for real-time predictions. Results show that the system
provides a real-time human activity recognition with an accuracy of 86%.
Keywords: Machine learning · Deep learning · HAR · Embedded
ANN · LSTM-RNN · Raspberry PI · Python
1 Introduction
Human activity recognition (HAR) refers to the automatic detection of various
physical activities performed by people in their daily lives [1]. Activity recognition can be achieved by exploiting information retrieved from sensors such as
Supported by German Academic Exchange Service, CRNS and ReDCAD.
c
The Author(s) 2020
M. Jmaiel et al. (Eds.): ICOST 2020, LNCS 12157, pp. 375–385, 2020.
https://doi.org/10.1007/978-3-030-51517-1_34