Next Article in Journal
Design and Optimization of a High-Time-Resolution Magnetic Plasma Analyzer (MPA)
Next Article in Special Issue
A Deep Neural Network Model for Speaker Identification
Previous Article in Journal
Decomposition Methods for Machine Learning with Small, Incomplete or Noisy Datasets
Previous Article in Special Issue
Deep Learning Models Compression for Agricultural Plants
Article

Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks

Centre for Research and Technology—Hellas, Information Technologies Institute, 6th km Charilaou-Thermi Rd, 57001 Thessaloniki, Greece
*
Author to whom correspondence should be addressed.
Appl. Sci. 2020, 10(23), 8482; https://doi.org/10.3390/app10238482
Received: 20 October 2020 / Revised: 24 November 2020 / Accepted: 24 November 2020 / Published: 27 November 2020
(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)
Given the ubiquity of mobile devices, understanding the context of human activity with non-intrusive solutions is of great value. A novel deep neural network model is proposed, which combines feature extraction and convolutional layers, able to recognize human physical activity in real-time from tri-axial accelerometer data when run on a mobile device. It uses a two-layer convolutional neural network to extract local features, which are combined with 40 statistical features and are fed to a fully-connected layer. It improves the classification performance, while it takes up 5–8 times less storage space and outputs more than double the throughput of the current state-of-the-art user-independent implementation on the Wireless Sensor Data Mining (WISDM) dataset. It achieves 94.18% classification accuracy on a 10-fold user-independent cross-validation of the WISDM dataset. The model is further tested on the Actitracker dataset, achieving 79.12% accuracy, while the size and throughput of the model are evaluated on a mobile device. View Full-Text
Keywords: activity recognition; convolutional neural networks; deep learning; human activity recognition; mobile inference; time series classification; feature extraction; accelerometer data activity recognition; convolutional neural networks; deep learning; human activity recognition; mobile inference; time series classification; feature extraction; accelerometer data
Show Figures

Figure 1

MDPI and ACS Style

Peppas, K.; Tsolakis, A.C.; Krinidis, S.; Tzovaras, D. Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks. Appl. Sci. 2020, 10, 8482. https://doi.org/10.3390/app10238482

AMA Style

Peppas K, Tsolakis AC, Krinidis S, Tzovaras D. Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks. Applied Sciences. 2020; 10(23):8482. https://doi.org/10.3390/app10238482

Chicago/Turabian Style

Peppas, Konstantinos, Apostolos C. Tsolakis, Stelios Krinidis, and Dimitrios Tzovaras. 2020. "Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks" Applied Sciences 10, no. 23: 8482. https://doi.org/10.3390/app10238482

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop