A Deep Learning Approach to Human Activity Recognition Based on Single Accelerometer

In this paper, we propose an acceleration-based human activity recognition method using popular deep architecture, Convolution Neural Network (CNN). In particular, we construct a CNN model and modify the convolution kernel to adapt the characteristics of tri-axial acceleration signals. Also, for com...

Full description

Saved in:
Bibliographic Details
Published in2015 IEEE International Conference on Systems, Man, and Cybernetics pp. 1488 - 1492
Main Authors Yuqing Chen, Yang Xue
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2015
Subjects
Acceleration
Convolution
convolution kernel
deep architecture
Discrete cosine transforms
Feature extraction
human activity recognition
Kernel
Legged locomotion
Training
tri-axial acceleration signal
Online AccessGet full text
DOI10.1109/SMC.2015.263

Cover

More Information
Summary:In this paper, we propose an acceleration-based human activity recognition method using popular deep architecture, Convolution Neural Network (CNN). In particular, we construct a CNN model and modify the convolution kernel to adapt the characteristics of tri-axial acceleration signals. Also, for comparison, we use some widely used methods to accomplish the recognition task on the same dataset. The large dataset we constructed consists of 31688 samples from eight typical activities. The experiment results show that the CNN works well, which can reach an average accuracy of 93.8% without any feature extraction methods.
DOI:10.1109/SMC.2015.263