A Multi-radar Architecture for Human Activity Recognition in Indoor Kitchen Environments

This paper tackles the Human Activity Recognition (HAR) in the kitchen environment using radar as the sensing tool. The setup includes two Frequency Modulated Continuous Wave (FMCW) radars each mounted separately on the ceiling and the wall, respectively. The data collected from concurrently operate...

Full description

Saved in:
Bibliographic Details
Published inProceedings of the IEEE National Radar Conference (1996) pp. 1 - 6
Main Authors Gorji, A., Gielen, T., Bauduin, M., Sahli, H., Bourdoux, A.
Format Conference Proceeding
LanguageEnglish
Published IEEE 07.05.2021
Subjects
Activity recognition
Feature extraction
Pipelines
Radar
Radar detection
Radar tracking
Online AccessGet full text

Cover

More Information
Summary:This paper tackles the Human Activity Recognition (HAR) in the kitchen environment using radar as the sensing tool. The setup includes two Frequency Modulated Continuous Wave (FMCW) radars each mounted separately on the ceiling and the wall, respectively. The data collected from concurrently operated radars was used to evaluate the efficacy of the HAR. In addition, an indoor kitchen scenario in the presence of furniture is considered and the HAR is taken as a procedure to detect common cooking-related activities by a single human subject where a machine learning model is developed to address the HAR as a multi-class classification problem. Our experimental results show the superior performance of the proposed method in detecting kitchen activities, especially, when the features from both the radars are being fused in the central processor. 1
ISSN:2375-5318
DOI:10.1109/RadarConf2147009.2021.9455238