Pedestrian Navigation Activity Recognition Based on Segmentation Transformer

doi:10.1109/JIOT.2024.3394050

	Pedestrian Navigation Activity Recognition Based on Segmentation Transformer
	Wang, Qu 1,2; Tao, Zhi 1; Ning, Jiahui 1; Jiang, Zhuqing 1; Guo, Liangliang 3; Luo, Haiyong 4; Wang, Haiying 1; Men, Aidong 1; Cheng, Xiaofei 5; Zhang, Zhang 6
	2024-08-01
发表期刊	IEEE INTERNET OF THINGS JOURNAL
ISSN	2327-4662
卷号	11 期号:15 页码:26020-26032
摘要	In the context of the Internet of Things, utilizing the inherent inertial sensors in smartphones for human activity recognition (HAR) has garnered considerable attention owing to its wide-ranging applications. However, prevailing HAR approaches primarily treat activity identification as a single-label classification task, focusing solely on discerning pedestrian motion modes or device usage modes, while disregarding their interrelatedness. Additionally, HAR methods employing sliding windows encounter challenges associated with the multiclass window problem, wherein certain sample labels differ from the label assigned to the window. This article aims to address these issues. This article presents a novel approach for simultaneously recognizing pedestrian motion and device usage modes by utilizing the segmentation Transformer. The proposed joint recognition framework effectively annotates sensor data at each timestamp and achieves dense prediction of time-series data through the encoding and decoding of the annotated data. To optimize the utilization of information extracted from each Transformer layer, a global up-sampling decoder based on the pyramid attention module is introduced, enabling dense decoding of features obtained from each Transformer layer. We performed experiments on two publicly available data sets to comprehensively assess the effectiveness of the proposed methodology. The results demonstrate that our approach achieves an accuracy of 99.79% and a weighted F-score of 99.77%, surpassing the performance of existing state-of-the-art methods. Furthermore, we constructed heterogeneous data sets to validate the robustness of our method. The extensive experimental findings indicate that the joint recognition framework effectively uncovers the inherent correlations between pedestrian motion and device usage modes, leading to enhanced accuracy in recognition and addressing the challenges posed by the multiclass window problem.
关键词	Feature extraction Pedestrians Transformers Hidden Markov models Human activity recognition Data mining Artificial intelligence for IoT dense sequence labeling human activity recognition (HAR) Internet of Things (IoT) multiclass window problem pedestrian navigation pedestrian navigation
DOI	10.1109/JIOT.2024.3394050
收录类别	SCI
语种	英语
资助项目	Joint Research Fund for Beijing Natural Science Foundation ; Haidian Original Innovation[L232001] ; GuangDong Basic and Applied Basic Research Foundation[2024A1515011866] ; GuangDong Basic and Applied Basic Research Foundation[2024A1515011480] ; Central Guidance on Local Science and Technology Development Fund of ShanXi Province[YDZJSX20231D005] ; Central Guidance on Local Science and Technology Development Fund of ShanXi Province[YDZJSX2022B019] ; Central Guidance on Local Science and Technology Development Fund of ShanXi Province[YDZJSX20231B017] ; National Natural Science Foundation of China[62002026] ; National Natural Science Foundation of China[61872046] ; University of Science and Technology Beijing Young Faculty International Exchange and Development Program[QNXM20230016] ; Beijing Science and Technology Plan[Z231100005923025]
WOS研究方向	Computer Science ; Engineering ; Telecommunications
WOS类目	Computer Science, Information Systems ; Engineering, Electrical & Electronic ; Telecommunications
WOS记录号	WOS:001277988600042
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/39626
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Jiang, Zhuqing; Zhang, Zhang
作者单位	1.Univ Sci & Technol Beijing, Sch Automat Sci & Elect Engn, Beijing 100083, Peoples R China 2.Univ Sci & Technol Beijing, Shunde Innovat Sch, Foshan 528399, Peoples R China 3.Shanxi Informat Ind Technol Res Inst Co Ltd, Dept Energy & Control Engn, Taiyuan 030012, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing Key Lab Mobile Comp & Pervas Device, Beijing 100190, Peoples R China 5.Keppel Corp Ltd, Keppel Bay Tower, Singapore 639798, Singapore 6.China Elect Standardizat Inst, IoT Res Ctr, Beijing 100007, Peoples R China
推荐引用方式 GB/T 7714	Wang, Qu,Tao, Zhi,Ning, Jiahui,et al. Pedestrian Navigation Activity Recognition Based on Segmentation Transformer[J]. IEEE INTERNET OF THINGS JOURNAL,2024,11(15):26020-26032.
APA	Wang, Qu.,Tao, Zhi.,Ning, Jiahui.,Jiang, Zhuqing.,Guo, Liangliang.,...&Zhang, Zhang.(2024).Pedestrian Navigation Activity Recognition Based on Segmentation Transformer.IEEE INTERNET OF THINGS JOURNAL,11(15),26020-26032.
MLA	Wang, Qu,et al."Pedestrian Navigation Activity Recognition Based on Segmentation Transformer".IEEE INTERNET OF THINGS JOURNAL 11.15(2024):26020-26032.