Real-time occupancy estimation using environmental parameters

Mustafa K. Masood, Yeng Chai Soh, Victor W.-C. Chang

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

Abstract

An integral part of visualizing an air-conditioned space is to know its occupancy in real-time, in order to make intelligent control decisions about the operation of its Air Conditioning and Mechanical Ventilation (ACMV) system. The sensing mechanisms used in occupancy estimation such as cameras and wearable sensors are generally intrusive and expensive. Alternatively, the effect that occupants have on environmental parameters such as CO2, temperature, humidity and pressure can be utilized to extract information about the occupancy levels. Environmental sensors are relatively inexpensive and are non-intrusive. From these sensor data, we need to extract and select relevant features that may yield occupancy information. The filter model feature selection approach used in previous works compromises on the classification accuracy in order to limit the computational burden. An alternative is the wrapper model of feature selection, which uses the inference algorithm itself to search for the best features. It guarantees better classification accuracy but is computationally expensive, especially with slow iterative machine learning techniques such as the Artificial Neural Network (ANN) used in previous works. To address this problem, this work capitalizes on the fast learning speed of Extreme Learning Machines (ELM) to implement a wrapper model of feature selection. To the best of our knowledge, the use of the wrapper model in an occupancy estimation problem has not been documented. A comparison between the filter and wrapper model feature selection is made. The tracking accuracy was seen to have notably improved with the wrapper model. Also, it was demonstrated that the pressure data, which has not been used for occupancy estimation in previous works, is useful.

Original languageEnglish
Title of host publication2015 International Joint Conference on Neural Networks, IJCNN 2015
Subtitle of host publicationKillarney, Ireland, 12-17 July 2015
EditorsYoonsuck Choe
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages3545-3552
Number of pages8
Volume2015-September
ISBN (Electronic)9781479919604
ISBN (Print)9781479919611
DOIs
Publication statusPublished - 28 Sep 2015
EventIEEE International Joint Conference on Neural Networks 2015 - Killarney, Ireland
Duration: 12 Jul 201517 Jul 2015

Conference

ConferenceIEEE International Joint Conference on Neural Networks 2015
Abbreviated titleIJCNN 2015
CountryIreland
CityKillarney
Period12/07/1517/07/15

Keywords

  • ACMV
  • features
  • occupancy
  • wrapper model

Cite this

Masood, M. K., Soh, Y. C., & Chang, V. W-C. (2015). Real-time occupancy estimation using environmental parameters. In Y. Choe (Ed.), 2015 International Joint Conference on Neural Networks, IJCNN 2015: Killarney, Ireland, 12-17 July 2015 (Vol. 2015-September, pp. 3545-3552). [7280781] Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IJCNN.2015.7280781
Masood, Mustafa K. ; Soh, Yeng Chai ; Chang, Victor W.-C. / Real-time occupancy estimation using environmental parameters. 2015 International Joint Conference on Neural Networks, IJCNN 2015: Killarney, Ireland, 12-17 July 2015. editor / Yoonsuck Choe. Vol. 2015-September Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 3545-3552
@inproceedings{214baf40e929425b886d1bf72e4bb8be,
title = "Real-time occupancy estimation using environmental parameters",
abstract = "An integral part of visualizing an air-conditioned space is to know its occupancy in real-time, in order to make intelligent control decisions about the operation of its Air Conditioning and Mechanical Ventilation (ACMV) system. The sensing mechanisms used in occupancy estimation such as cameras and wearable sensors are generally intrusive and expensive. Alternatively, the effect that occupants have on environmental parameters such as CO2, temperature, humidity and pressure can be utilized to extract information about the occupancy levels. Environmental sensors are relatively inexpensive and are non-intrusive. From these sensor data, we need to extract and select relevant features that may yield occupancy information. The filter model feature selection approach used in previous works compromises on the classification accuracy in order to limit the computational burden. An alternative is the wrapper model of feature selection, which uses the inference algorithm itself to search for the best features. It guarantees better classification accuracy but is computationally expensive, especially with slow iterative machine learning techniques such as the Artificial Neural Network (ANN) used in previous works. To address this problem, this work capitalizes on the fast learning speed of Extreme Learning Machines (ELM) to implement a wrapper model of feature selection. To the best of our knowledge, the use of the wrapper model in an occupancy estimation problem has not been documented. A comparison between the filter and wrapper model feature selection is made. The tracking accuracy was seen to have notably improved with the wrapper model. Also, it was demonstrated that the pressure data, which has not been used for occupancy estimation in previous works, is useful.",
keywords = "ACMV, features, occupancy, wrapper model",
author = "Masood, {Mustafa K.} and Soh, {Yeng Chai} and Chang, {Victor W.-C.}",
year = "2015",
month = "9",
day = "28",
doi = "10.1109/IJCNN.2015.7280781",
language = "English",
isbn = "9781479919611",
volume = "2015-September",
pages = "3545--3552",
editor = "Choe, {Yoonsuck }",
booktitle = "2015 International Joint Conference on Neural Networks, IJCNN 2015",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Masood, MK, Soh, YC & Chang, VW-C 2015, Real-time occupancy estimation using environmental parameters. in Y Choe (ed.), 2015 International Joint Conference on Neural Networks, IJCNN 2015: Killarney, Ireland, 12-17 July 2015. vol. 2015-September, 7280781, IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 3545-3552, IEEE International Joint Conference on Neural Networks 2015, Killarney, Ireland, 12/07/15. https://doi.org/10.1109/IJCNN.2015.7280781

Real-time occupancy estimation using environmental parameters. / Masood, Mustafa K.; Soh, Yeng Chai; Chang, Victor W.-C.

2015 International Joint Conference on Neural Networks, IJCNN 2015: Killarney, Ireland, 12-17 July 2015. ed. / Yoonsuck Choe. Vol. 2015-September Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 3545-3552 7280781.

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

TY - GEN

T1 - Real-time occupancy estimation using environmental parameters

AU - Masood, Mustafa K.

AU - Soh, Yeng Chai

AU - Chang, Victor W.-C.

PY - 2015/9/28

Y1 - 2015/9/28

N2 - An integral part of visualizing an air-conditioned space is to know its occupancy in real-time, in order to make intelligent control decisions about the operation of its Air Conditioning and Mechanical Ventilation (ACMV) system. The sensing mechanisms used in occupancy estimation such as cameras and wearable sensors are generally intrusive and expensive. Alternatively, the effect that occupants have on environmental parameters such as CO2, temperature, humidity and pressure can be utilized to extract information about the occupancy levels. Environmental sensors are relatively inexpensive and are non-intrusive. From these sensor data, we need to extract and select relevant features that may yield occupancy information. The filter model feature selection approach used in previous works compromises on the classification accuracy in order to limit the computational burden. An alternative is the wrapper model of feature selection, which uses the inference algorithm itself to search for the best features. It guarantees better classification accuracy but is computationally expensive, especially with slow iterative machine learning techniques such as the Artificial Neural Network (ANN) used in previous works. To address this problem, this work capitalizes on the fast learning speed of Extreme Learning Machines (ELM) to implement a wrapper model of feature selection. To the best of our knowledge, the use of the wrapper model in an occupancy estimation problem has not been documented. A comparison between the filter and wrapper model feature selection is made. The tracking accuracy was seen to have notably improved with the wrapper model. Also, it was demonstrated that the pressure data, which has not been used for occupancy estimation in previous works, is useful.

AB - An integral part of visualizing an air-conditioned space is to know its occupancy in real-time, in order to make intelligent control decisions about the operation of its Air Conditioning and Mechanical Ventilation (ACMV) system. The sensing mechanisms used in occupancy estimation such as cameras and wearable sensors are generally intrusive and expensive. Alternatively, the effect that occupants have on environmental parameters such as CO2, temperature, humidity and pressure can be utilized to extract information about the occupancy levels. Environmental sensors are relatively inexpensive and are non-intrusive. From these sensor data, we need to extract and select relevant features that may yield occupancy information. The filter model feature selection approach used in previous works compromises on the classification accuracy in order to limit the computational burden. An alternative is the wrapper model of feature selection, which uses the inference algorithm itself to search for the best features. It guarantees better classification accuracy but is computationally expensive, especially with slow iterative machine learning techniques such as the Artificial Neural Network (ANN) used in previous works. To address this problem, this work capitalizes on the fast learning speed of Extreme Learning Machines (ELM) to implement a wrapper model of feature selection. To the best of our knowledge, the use of the wrapper model in an occupancy estimation problem has not been documented. A comparison between the filter and wrapper model feature selection is made. The tracking accuracy was seen to have notably improved with the wrapper model. Also, it was demonstrated that the pressure data, which has not been used for occupancy estimation in previous works, is useful.

KW - ACMV

KW - features

KW - occupancy

KW - wrapper model

UR - http://www.scopus.com/inward/record.url?scp=84951141932&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2015.7280781

DO - 10.1109/IJCNN.2015.7280781

M3 - Conference Paper

SN - 9781479919611

VL - 2015-September

SP - 3545

EP - 3552

BT - 2015 International Joint Conference on Neural Networks, IJCNN 2015

A2 - Choe, Yoonsuck

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

ER -

Masood MK, Soh YC, Chang VW-C. Real-time occupancy estimation using environmental parameters. In Choe Y, editor, 2015 International Joint Conference on Neural Networks, IJCNN 2015: Killarney, Ireland, 12-17 July 2015. Vol. 2015-September. Piscataway NJ USA: IEEE, Institute of Electrical and Electronics Engineers. 2015. p. 3545-3552. 7280781 https://doi.org/10.1109/IJCNN.2015.7280781