Multi-ResAtt: Multilevel Residual Network With Attention for Human Activity Recognition Using Wearable Sensors

被引:91
作者
Al-qaness, Mohammed A. A. [1 ]
Dahou, Abdelghani [2 ,3 ]
Abd Elaziz, Mohamed [4 ,5 ,6 ]
Helmi, A. M. [7 ,8 ]
机构
[1] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan 430079, Peoples R China
[2] Univ Ahmed DRAIA, Fac Sci & Technol, Lab Dev Durable & Informat LDDI Lab, Adrar 01000, Algeria
[3] Univ Ahmed DRAIA, Dept Math & Comp Sci, Adrar 01000, Algeria
[4] Galala Univ, Fac Comp Sci & Engn, Suez 435611, Egypt
[5] Zagazig Univ, Dept Math, Fac Sci, Zagazig 44519, Egypt
[6] Ajman Univ, Artificial Intelligence Res Ctr, Ajman 346, U Arab Emirates
[7] Zagazig Univ, Fac Engn, Dept Comp & Syst Engn, Zagazig 44519, Egypt
[8] Buraydah Private Coll, Coll Engn & Informat Technol, Buraydah 51418, Saudi Arabia
基金
中国国家自然科学基金;
关键词
Sensors; Deep learning; Feature extraction; Convolutional neural networks; Activity recognition; Residual neural networks; Informatics; Deep learning (DL); human activity recognition; Industry; 5; 0; Internet of Things (IoT); recurrent neural network (RNN); wearable sensors;
D O I
10.1109/TII.2022.3165875
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Human activity recognition (HAR) applications have received much attention due to their necessary implementations in various domains, including Industry 5.0 applications such as smart homes, e-health, and various Internet of Things applications. Deep learning (DL) techniques have shown impressive performance in different classification tasks, including HAR. Accordingly, in this article, we develop a comprehensive HAR system based on a novel DL architecture called Multi-ResAtt (multilevel residual network with attention). This model incorporates initial blocks and residual modules aligned in parallel. Multi-ResAtt learns data representations on the inertial measurement units level. Multi-ResAtt integrates a recurrent neural network with attention to extract time-series features and perform activity recognition. We consider complex human activities collected from wearable sensors to evaluate the Multi-ResAtt using three public datasets, Opportunity; UniMiB-SHAR; and PAMAP2. Additionally, we compared the proposed Multi-ResAtt to several DL models and existing HAR systems, and it achieved significant performance.
引用
收藏
页码:144 / 152
页数:9
相关论文
共 50 条
[41]   An Efficient Hierarchical Multiscale and Multidimensional Feature Adaptive Fusion Network for Human Activity Recognition Using Wearable Sensors [J].
Li, Xinya ;
Xu, Hongji ;
Wang, Yang ;
Zeng, Jiaqi ;
Li, Yiran ;
Li, Xiaoman ;
Ai, Wentao ;
Zheng, Hao ;
Duan, Yupeng .
IEEE INTERNET OF THINGS JOURNAL, 2025, 12 (06) :6492-6505
[42]   Revolutionizing health monitoring: Integrating transformer models with multi-head attention for precise human activity recognition using wearable devices [J].
Muniasamy, Anandhavalli .
TECHNOLOGY AND HEALTH CARE, 2025, 33 (01) :395-409
[43]   Short-time Activity Recognition With Wearable Sensors Using Convolutional Neural Network [J].
Sheng, Min ;
Jiang, Jing ;
Su, Benyue ;
Tang, Qingfeng ;
Yahya, Ali Abdullah ;
Wang, Guangjun .
PROCEEDINGS VRCAI 2016: 15TH ACM SIGGRAPH CONFERENCE ON VIRTUAL-REALITY CONTINUUM AND ITS APPLICATIONS IN INDUSTRY, 2016, :413-416
[44]   Multi-Branch Attention-Based Grouped Convolution Network for Human Activity Recognition Using Inertial Sensors [J].
Li, Yong ;
Wang, Luping ;
Liu, Fen .
ELECTRONICS, 2022, 11 (16)
[45]   Wearable sensors for human activity recognition based on a self-attention CNN-BiLSTM model [J].
Guo, Huafeng ;
Xiang, Changcheng ;
Chen, Shiqiang .
SENSOR REVIEW, 2023, 43 (5/6) :347-358
[46]   GLULA: Linear attention-based model for efficient human activity recognition from wearable sensors [J].
Bolatov, Aldiyar ;
Yessenbayeva, Aigerim ;
Yazici, Adnan .
WEARABLE TECHNOLOGIES, 2024, 5
[47]   A Multi-Section Hierarchical Deep Neural Network Model for Time Series Classification: Applied to Wearable Sensor-Based Human Activity Recognition [J].
Ghorrati, Zahra ;
Esmaeili, Ahmad ;
Matson, Eric T. .
IEEE ACCESS, 2024, 12 :137851-137869
[48]   Device Position-Independent Human Activity Recognition with Wearable Sensors Using Deep Neural Networks [J].
Mekruksavanich, Sakorn ;
Jitpattanakul, Anuchit .
APPLIED SCIENCES-BASEL, 2024, 14 (05)
[49]   Weakly Supervised Human Activity Recognition From Wearable Sensors by Recurrent Attention Learning [J].
He, Jun ;
Zhang, Qian ;
Wang, Liqun ;
Pei, Ling .
IEEE SENSORS JOURNAL, 2019, 19 (06) :2287-2297
[50]   Sensors-based Human Activity Recognition with Convolutional Neural Network and Attention Mechanism [J].
Zhang, Wenbo ;
Zhu, Tao ;
Yang, Congmin ;
Xiao, Jiyi ;
Ning, Huansheng .
PROCEEDINGS OF 2020 IEEE 11TH INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING AND SERVICE SCIENCE (ICSESS 2020), 2020, :158-162