Occupancy detection, including presence detection and head count, as one of the fast growing areas plays an important role in providing safety, comfort and reducing energy consumption both in residential and commercial setups. The focus of this study is proposing affordable strategies to increase occupancy detection performance in realistic scenarios using only audio signal collected from the environment. We use approximately 100-hour of audio data in residential and commercial environments to analyze and evaluate our setup. In this study, we take advantage of developments in feature selection methods to choose the most relevant audio features for the task. Attribute and error vs. human activity analysis are also performed to gain a better understanding of the environmental sounds and possible solutions to enhance the performance. Experimental results confirm the effectiveness of audio sensor for occupancy detection using a cost effective system with presence detection accuracy of 96% and 99%, and the head count accuracy of 70% and 95% for the residential and commercial setups, respectively.
Cite as: Ghaffarzadegan, S., Reiss, A., Ruhs, M., Duerichen, R., Feng, Z. (2017) Occupancy Detection in Commercial and Residential Environments Using Audio Signal. Proc. Interspeech 2017, 3802-3806, doi: 10.21437/Interspeech.2017-524
@inproceedings{ghaffarzadegan17_interspeech,
author={Shabnam Ghaffarzadegan and Attila Reiss and Mirko Ruhs and Robert Duerichen and Zhe Feng},
title={{Occupancy Detection in Commercial and Residential Environments Using Audio Signal}},
year=2017,
booktitle={Proc. Interspeech 2017},
pages={3802--3806},
doi={10.21437/Interspeech.2017-524}
}