TY - JOUR
T1 - Multi-resident activity recognition in a smart home using RGB activity image and DCNN
AU - Tan, Tan Hsu
AU - Gochoo, Munkhjargal
AU - Huang, Shih Chia
AU - Liu, Yi Hung
AU - Liu, Shing Hong
AU - Huang, Yun Fa
N1 - Funding Information:
Manuscript received July 27, 2018; accepted August 18, 2018. Date of publication August 23, 2018; date of current version November 13, 2018. This work was supported by the Ministry of Science and Technology of the Republic of China (Taiwan) under Contract MOST 106-2218-E-027-017 and Contract MOST 107-2218-E-027-011. The associate editor coordinating the review of this paper and approving it for publication was Prof. Dongsoo Har. (Corresponding author: Shih-Chia Huang.) T.-H. Tan is with the Department of Electrical Engineering, National Taipei University of Technology, Taipei 10608, Taiwan (e-mail: thtan@ntut.edu.tw).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/12/1
Y1 - 2018/12/1
N2 - In the last decade, unobtrusive (device-free and non-privacy invasive) recognition of activities of daily living for an individual in a smart home has been studied by many researchers. However, the unobtrusive recognition of multi-resident activities in a smart home is hardly studied. We propose a novel RGB activity image-based DCNN classifier for the unobtrusive recognition of the multi-resident activities (BedtoToilet, Bed, Breakfast, Lunch, Leavehome, Laundry, Dinner, Nightwandering, R2work, and R1medicine) using Cairo open data set provided by the CASAS Project. The open data set is collected by environmental sensors (PIR and temperature sensors) in Cairo testbed, while an adult couple with a dog was living for 55 days. The data set is preprocessed with activity segmentation, sliding window, and RGB activity image conversion steps. The experimental results demonstrate that our classifier has the highest total accuracy of 95.2% among the previously developed machine learning classifiers that employed the same data set. Moreover, the proposed RGB activity image was proven to be helpful for increasing the recognition rate. Therefore, we conclude that the proposed DCNN classifier is a useful tool for the unobtrusive recognition of the multi-resident activity in a home.
AB - In the last decade, unobtrusive (device-free and non-privacy invasive) recognition of activities of daily living for an individual in a smart home has been studied by many researchers. However, the unobtrusive recognition of multi-resident activities in a smart home is hardly studied. We propose a novel RGB activity image-based DCNN classifier for the unobtrusive recognition of the multi-resident activities (BedtoToilet, Bed, Breakfast, Lunch, Leavehome, Laundry, Dinner, Nightwandering, R2work, and R1medicine) using Cairo open data set provided by the CASAS Project. The open data set is collected by environmental sensors (PIR and temperature sensors) in Cairo testbed, while an adult couple with a dog was living for 55 days. The data set is preprocessed with activity segmentation, sliding window, and RGB activity image conversion steps. The experimental results demonstrate that our classifier has the highest total accuracy of 95.2% among the previously developed machine learning classifiers that employed the same data set. Moreover, the proposed RGB activity image was proven to be helpful for increasing the recognition rate. Therefore, we conclude that the proposed DCNN classifier is a useful tool for the unobtrusive recognition of the multi-resident activity in a home.
KW - Unobtrusive
KW - activity recognition
KW - convolutional neural networks
KW - deep learning
KW - wandering detection
UR - http://www.scopus.com/inward/record.url?scp=85054700866&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85054700866&partnerID=8YFLogxK
U2 - 10.1109/JSEN.2018.2866806
DO - 10.1109/JSEN.2018.2866806
M3 - Article
AN - SCOPUS:85054700866
SN - 1530-437X
VL - 18
SP - 9718
EP - 9727
JO - IEEE Sensors Journal
JF - IEEE Sensors Journal
IS - 23
M1 - 8444711
ER -