The Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network

doi:10.1109/JIOT.2023.3234053

CORC > 自动化研究所 > 中国科学院自动化研究所 > 精密感知与控制研究中心 > 人工智能与机器学习

	The Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network
	Yuan, Ruiwen 1,2; Wang, Junping 1,2
刊名	IEEE INTERNET OF THINGS JOURNAL
	2023-06-01
卷号	10 期号:11 页码:9398-9410
关键词	Feature extraction Sensors Human activity recognition Internet of Things Data models Sensor phenomena and characterization Data mining Activity feature extraction deep learning human activity recognition (HAR) semisupervised learning
ISSN号	2327-4662
DOI	10.1109/JIOT.2023.3234053
通讯作者	Yuan, Ruiwen(yuanruiwen2021@ia.ac.cn)
英文摘要	With advances in spatial-temporal Internet of Things (IoT) technologies, human activity recognition (HAR) has played a major role in human safety and medical health. Recently, most works focus on activity deep feature extraction, offering promising alternatives to manually engineered features. However, how to extract the effective and distinguishable continuity activity features and meanwhile reduce the heavy dependence on labels still remains the key challenge for HAR. This article proposes the human continuity activity semisupervised recognizing method in multiview IoT network scenarios. Our innovation combines supervised activity feature extraction with unsupervised encoder-decoder modules, which can capture continuity activity features from sensor data streams. To be more specific, our work applies a convolutional neural network (CNN) to capture the local dependence of sensor data and designs an encoder-decoder architecture to extract temporal features in an unsupervised manner. Then, we fuse these two features to recognize activities and train them with manual labels, thereby refining the temporal feature extraction and training CNN module. Experiments on five public data sets demonstrate the exceptional performance of our proposed method, which can achieve a higher recognition accuracy on almost all the data sets and is more robust and adaptive among different data sets.
资助项目	National Key Research and Development Program of China[2022YFF0903304] ; National Natural Science Foundation of China[92167109] ; Dadu River Cascade Hydropower Station Safety Early Warning Project
WOS关键词	HUMAN ACTIVITY RECOGNITION ; WEARABLE SENSOR ; LEARNING APPROACH ; FEATURE FUSION
WOS研究方向	Computer Science ; Engineering ; Telecommunications
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000991733300015
资助机构	National Key Research and Development Program of China ; National Natural Science Foundation of China ; Dadu River Cascade Hydropower Station Safety Early Warning Project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/53537]
专题	精密感知与控制研究中心_人工智能与机器学习
通讯作者	Yuan, Ruiwen
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 101408, Peoples R China 2.Chinese Acad Sci, Inst Automat, Res Ctr Precis Sensing & Control, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Yuan, Ruiwen,Wang, Junping. The Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network[J]. IEEE INTERNET OF THINGS JOURNAL,2023,10(11):9398-9410.
APA	Yuan, Ruiwen,&Wang, Junping.(2023).The Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network.IEEE INTERNET OF THINGS JOURNAL,10(11),9398-9410.
MLA	Yuan, Ruiwen,et al."The Human Continuity Activity Semisupervised Recognizing Model for Multiview IoT Network".IEEE INTERNET OF THINGS JOURNAL 10.11(2023):9398-9410.