TY - JOUR
T1 - Novel IoT-Based Privacy-Preserving Yoga Posture Recognition System Using Low-Resolution Infrared Sensors and Deep Learning
AU - Gochoo, Munkhjargal
AU - Tan, Tan Hsu
AU - Huang, Shih Chia
AU - Batjargal, Tsedevdorj
AU - Hsieh, Jun Wei
AU - Alnajjar, Fady S.
AU - Chen, Yung Fu
N1 - Funding Information:
Manuscript received October 15, 2018; revised March 27, 2019; accepted May 2, 2019. Date of publication May 6, 2019; date of current version July 31, 2019. This work was supported in part by the Ministry of Science and Technology, Taiwan, under Contract MOST 107-2221-E-027-106, Contract MOST 107-2218-E-027-011, and Contract MOST 108-2634-F-008-002. An earlier version of this paper was presented at SMC2018 and published in the conference proceedings. (Corresponding author: Shih-Chia Huang.) M. Gochoo is with the Department of Electronic Engineering, National Taipei University of Technology, Taipei 10608, Taiwan, and also with the School of Information and Communication Technology, Mongolian University of Science and Technology, Ulaanbaatar 13341, Mongolia (e-mail: g.munkhjargal@must.edu.mn).
Funding Information:
This work was supported in part by the Ministry of Science and Technology, Taiwan, under Contract MOST 107-2221-E-027-106, Contract MOST 107-2218-E-027-011
Publisher Copyright:
© 2014 IEEE.
PY - 2019/8
Y1 - 2019/8
N2 - In recent years, the number of yoga practitioners has been drastically increased and there are more men and older people practice yoga than ever before. Internet of Things (IoT)-based yoga training system is needed for those who want to practice yoga at home. Some studies have proposed RGB/Kinect camera-based or wearable device-based yoga posture recognition methods with a high accuracy; however, the former has a privacy issue and the latter is impractical in the long-term application. Thus, this paper proposes an IoT-based privacy-preserving yoga posture recognition system employing a deep convolutional neural network (DCNN) and a low-resolution infrared sensor-based wireless sensor network (WSN). The WSN has three nodes ( {x} , {y} , and {z} -axes) where each integrates 8\times 8 pixels' thermal sensor module and a Wi-Fi module for connecting the deep learning server. We invited 18 volunteers to perform 26 yoga postures for two sessions each lasted for 20 s. First, recorded sessions are saved as.csv files, then preprocessed and converted to grayscale posture images. Totally, 93 200 posture images are employed for the validation of the proposed DCNN models. The tenfold cross-validation results revealed that F1-scores of the models trained with xyz (all 3-axes) and {y} (only {y} -axis) posture images were 0.9989 and 0.9854, respectively. An average latency for a single posture image classification on the server was 107 ms. Thus, we conclude that the proposed IoT-based yoga posture recognition system has a great potential in the privacy-preserving yoga training system.
AB - In recent years, the number of yoga practitioners has been drastically increased and there are more men and older people practice yoga than ever before. Internet of Things (IoT)-based yoga training system is needed for those who want to practice yoga at home. Some studies have proposed RGB/Kinect camera-based or wearable device-based yoga posture recognition methods with a high accuracy; however, the former has a privacy issue and the latter is impractical in the long-term application. Thus, this paper proposes an IoT-based privacy-preserving yoga posture recognition system employing a deep convolutional neural network (DCNN) and a low-resolution infrared sensor-based wireless sensor network (WSN). The WSN has three nodes ( {x} , {y} , and {z} -axes) where each integrates 8\times 8 pixels' thermal sensor module and a Wi-Fi module for connecting the deep learning server. We invited 18 volunteers to perform 26 yoga postures for two sessions each lasted for 20 s. First, recorded sessions are saved as.csv files, then preprocessed and converted to grayscale posture images. Totally, 93 200 posture images are employed for the validation of the proposed DCNN models. The tenfold cross-validation results revealed that F1-scores of the models trained with xyz (all 3-axes) and {y} (only {y} -axis) posture images were 0.9989 and 0.9854, respectively. An average latency for a single posture image classification on the server was 107 ms. Thus, we conclude that the proposed IoT-based yoga posture recognition system has a great potential in the privacy-preserving yoga training system.
KW - CNN
KW - device-free
KW - infrared
KW - privacy-preserving
KW - yoga posture recognition
UR - http://www.scopus.com/inward/record.url?scp=85070208239&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85070208239&partnerID=8YFLogxK
U2 - 10.1109/JIOT.2019.2915095
DO - 10.1109/JIOT.2019.2915095
M3 - Article
AN - SCOPUS:85070208239
SN - 2327-4662
VL - 6
SP - 7192
EP - 7200
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
IS - 4
M1 - 8707064
ER -