مدل ترکیبی مبتنی بر ترنسفورمر برای شناسایی فعالیت‌های انسانی براساس داده‌های حسگرهای محیطی خانه‌های هوشمند

نوع مقاله : مقاله پژوهشی

نویسندگان

1 ، دانشکده مهندسی برق و کامپیوتر، دانشگاه شهاب دانش، قم، ایران

2 دانشکده مهندسی برق و کامپیوتر، دانشگاه صنعتی قم، قم، ایران

10.22091/jemsc.2025.12363.1257

چکیده

با گسترش روزافزون خانه‌های هوشمند، شناسایی دقیق و خودکار فعالیت‌های انسانی به یکی از چالش‌های کلیدی در حوزه هوش مصنوعی و اینترنت اشیا تبدیل شده است. این فناوری در حوزه‌هایی نظیر مراقبت از سالمندان، نظارت بر سلامت وارتقای امنیت خانه‌های هوشمند کاربردهای حیاتی دارد. در این پژوهش، یک روش ترکیبی نوین مبتنی بر یادگیری عمیق برای شناسایی فعالیت‌های انسانی معرفی می‌شود که از مدل‌های ترنسفورمر و واحد بازگشتی دروازه‌ای بهره می‌گیرد. مدل ترنسفورمر با مکانیسم توجه چندسری، توانایی تحلیل روابط طولانی‌مدت میان داده‌های حسگر را دارد و الگوهای رفتاری را با دقت بیشتری شناسایی می‌کندو واحد بازگشتی دروازه‌ای به دلیل توانایی‌اش در یادگیری الگوهای زمانی به بهبود دقت شناسایی فعالیت‌ها کمک شایانی می‌کند. نتایج ارزیابی‌ها نشان می‌دهد که مدل در مجموعه داده های آروبا به دقت 95.19٪ و میلان 80.01٪ دست یافته، که بیانگر توانایی بالای مدل پیشنهادی در تعمیم‌پذیری و شناسایی الگوهای رفتاری است. همچنین، در مقایسات انجام شده با روش‌های مشابه، مدل پیشنهادی عملکرد چشمگیری در بهبود شناسایی فعالیت‌های انسانی از خود نشان داده است.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

A Transformer-Based Hybrid Model for Human Activity Recognition Using Smart Home Environmental Sensor Data

نویسندگان [English]

  • Ronak Fatahi 1
  • Fatemeh Sadat Lesani 2
1 Faculty of Electrical and Computer Engineering, Shahab Danesh University, Qom, Iran
2 Department of electrical and Computer Engineering, Qom University of Technology, Qom, Iran
چکیده [English]

With the rapid expansion of smart homes, accurate and automatic recognition of human activities has become one of the key challenges in the fields of artificial intelligence and the Internet of Things. This technology has vital applications in areas such as elderly care, health monitoring, and enhancing the security of smart homes. In this research, a deep learning-based hybrid approach for human activity recognition is introduced, which utilizes transformer models and gated recurrent units. The transformer model, with its multi-head attention mechanism, has the ability to analyze long-term relationships among sensor data and identify behavioral patterns with higher precision. The gated recurrent unit, due to its capability in learning temporal patterns, significantly contributes to improving the accuracy of activity recognition.Evaluation results show that the model achieved an accuracy of 95.19% on the Aruba dataset and 89.01% on the Milan dataset, indicating the high generalizability and pattern recognition ability of the proposed model. Furthermore, compared to similar methods, the proposed model has demonstrated a remarkable performance in improving human activity recognition.

کلیدواژه‌ها [English]

  • Activity Recognition
  • Deep Learning
  • Smart Home
  • Transformer
  • GRU

مراجع

Alsarhan, T., Alawneh, L., Al-Zinati, M., & Al-Ayyoub, M. (2019, 27-30 Oct. 2019). Bidirectional Gated Recurrent Units For Human Activity Recognition Using Accelerometer Data. 2019 IEEE SENSORS,
Anbazhagan, K., Swamy, G., Janani, R., & Farakte, A. (2024, 22-23 March 2024). Deep Learning based Human Activity Recognition in Smart Home. 2024 4th International Conference on Data Engineering and Communication Systems (ICDECS),
Chen, D., Yongchareon, S., Lai, E. M.-K., Yu, J., Sheng, Q. Z., & Li, Y. (2022). Transformer with bidirectional GRU for nonintrusive, sensor-based activity recognition in a multiresident environment. IEEE Internet of Things Journal, 9(23), 23716-23727. https://doi.org/https://doi.org/10.1109/JIOT.2022.3190307
Chen, J., Jiang, D., & Zhang, Y. (2019). A hierarchical bidirectional GRU model with attention for EEG-based emotion classification. IEEE Access, 7, 118530-118540. https://doi.org/https://doi.org/10.1109/ACCESS.2019.2936817
Chitty-Venkata, K. T., Mittal, S., Emani, M., Vishwanath, V., & Somani, A. K. (2023). A survey of techniques for optimizing transformer inference. Journal of Systems Architecture, 102990. https://doi.org/https://doi.org/10.1016/j.sysarc.2023.102990
Choudhury, N. A., & Soni, B. (2023). An Adaptive Batch Size based-CNN-LSTM Framework for Human Activity Recognition in Uncontrolled Environment. IEEE Transactions on Industrial Informatics. https://doi.org/https://doi.org/10.1109/TII.2022.3229522
Cook, D. J. (2010). Learning Setting-Generalized Activity Models for Smart Spaces. IEEE Intell Syst, 2010(99), 1. https://doi.org/10.1109/mis.2010.112
Tavakkoli-Moghaddam, R., Akbari, A. H., Tanhaeean, M., Moghdani, R., Gholian-Jouybari, F., & Hajiaghaei-Keshteli, M. (2024). Multi-objective boxing match algorithm for multi-objective optimization problems. Expert Systems with Applications, 239, 122394. https://doi.org/10.1016/j.eswa.2023.122394
Yavari, M., Marvi, M., & Akbari, A. H. (2020). Semi-permutation-based genetic algorithm for order acceptance and scheduling in two-stage assembly problem. Neural Computing and Applications, 32, 2989-3003. https://doi.org/10.1007/s00521-019-04027-w
Tanhaeean, M., Tavakkoli-Moghaddam, R., & Akbari, A. H. (2022). Boxing match algorithm: A new meta-heuristic algorithm. Soft Computing, 26(24), 13277-13299. https://doi.org/10.1007/s00500-022-07518-6
Hussain, A., Hussain, T., Ullah, W., & Baik, S. W. (2022). Vision Transformer and Deep Sequence Learning for Human Activity Recognition in Surveillance Videos. Computational Intelligence and Neuroscience, 2022, 3454167. https://doi.org/10.1155/2022/3454167
Jiang, L., Wu, M., Che, L., Xu, X., Mu, Y., & Wu, Y. (2023). Continuous Human Motion Recognition Based on FMCW Radar and Transformer. Journal of Sensors, 2023. https://doi.org/https://doi.org/10.1155/2023/2951812
Kumar, P., Chauhan, S., & Awasthi, L. K. (2024). Human Activity Recognition (HAR) Using Deep Learning: Review, Methodologies, Progress and Future Research Directions. Archives of Computational Methods in Engineering, 31(1), 179-219. https://doi.org/https://doi.org/10.1007/s11831-023-09986-x
Kwapisz, J. R., Weiss, G. M., & Moore, S. A. (2011). Activity recognition using cell phone accelerometers. ACM SigKDD Explorations Newsletter, 12(2), 74-82. https://doi.org/https://doi.org/10.1145/1964897.1964918
Le, T.-H., Tran, T.-H., & Pham, C. (2022). Human action recognition from inertial sensors with Transformer. 2022 International Conference on Multimedia Analysis and Pattern Recognition (MAPR),
LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. nature, 521(7553), 436-444. https://doi.org/10.1038/nature14539
Lee, T.-H., Kim, H., & Lee, D. (2023). Transformer based Early Classification for Real-time Human Activity Recognition in Smart Homes. Proceedings of the 38th ACM/SIGAPP Symposium on Applied Computing,
Lesani, F. S., & Fatahi, R. (2024). A Review on Transformer-Based Methods for Human Activity Recognition. International Journal of Web Research, 7(4), 81-100. https://doi.org/10.22133/ijwr.2024.485291.1244
Liu, B., & Fang, S. (2023). Multi-level wavelet network based on CNN-Transformer hybrid attention for single image deraining. Neural Computing and Applications, 35(30), 22387-22404. https://doi.org/10.1007/s00521-023-08899-x
Liu, Y., Huang, W., Jiang, S., Zhao, B., Wang, S., Wang, S., & Zhang, Y. (2023). TransTM: A device-free method based on time-streaming multiscale transformer for human activity recognition. Defence Technology. https://doi.org/https://doi.org/10.1016/j.dt.2023.02.021
Mohsen, S. (2023). Recognition of human activity using GRU deep learning algorithm. Multimedia Tools and Applications, 82(30), 47733-47749. https://doi.org/10.1007/s11042-023-15571-y
Pan, J., Hu, Z., Yin, S., & Li, M. (2022). GRU with dual attentions for sensor-based human activity recognition. Electronics, 11(11), 1797.
Pareek, G., Nigam, S., & Singh, R. (2024). Modeling transformer architecture with attention layer for human activity recognition. Neural Computing and Applications. https://doi.org/10.1007/s00521-023-09362-7
Pramanik, R., Sikdar, R., & Sarkar, R. (2023). Transformer-based deep reverse attention network for multi-sensory human activity recognition. Engineering Applications of Artificial Intelligence, 122, 106150. https://doi.org/https://doi.org/10.1016/j.engappai.2023.106150
Saidani, O., Alsafyani, M., Alroobaea, R., Alturki, N., Jahangir, R., & Menzli, L. J. (2023). An Efficient Human Activity Recognition using Hybrid Features and Transformer Model. IEEE Access. https://doi.org/https://doi.org/10.1109/ACCESS.2023.3314492
Sharifi-Renani, M., Mahoor, M. H., & Clary, C. W. (2023). BioMAT: An Open-Source Biomechanics Multi-Activity Transformer for Joint Kinematic Predictions Using Wearable Sensors. Sensors, 23(13), 5778. https://doi.org/https://doi.org/10.3390/s23135778
Shavit, Y., & Klein, I. (2021). Boosting inertial-based human activity recognition with transformers. IEEE Access, 9, 53540-53547. https://doi.org/https://doi.org/10.1109/ACCESS.2021.3070646
Sunil, A., Sheth, M. H., & Shreyas, E. (2021). Usual and unusual human activity recognition in video using deep learning and artificial intelligence for security applications. 2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT),
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
Wan, J., O’grady, M. J., & O’Hare, G. M. (2015). Dynamic sensor event segmentation for real-time activity recognition in a smart home context. Personal and Ubiquitous Computing, 19, 287-301. https://doi.org/https://doi.org/10.1007/s00779-014-0824-x
Wang, J., Chen, Y., Hao, S., Peng, X., & Hu, L. (2019). Deep learning for sensor-based activity recognition: A survey. Pattern recognition letters, 119, 3-11. https://doi.org/https://doi.org/10.1016/j.patrec.2018.02.010
Wang, L., Zhang, Z., Wei, L., & Zhou, Y. (2024). CNN-GRU-Transformer Human Activity Recognition Model Based on Feature Fusion. 2024 6th International Conference on Frontier Technologies of Information and Computer (ICFTIC),
Wojek, C., Dorkó, G., Schulz, A., & Schiele, B. (2008). Sliding-windows for rapid object class localization: A parallel technique. Joint Pattern Recognition Symposium, 
Akbari, A. H., & Jafari, M. (2025). Development of a Deep Reinforcement Learning Algorithm in a Dynamic Cellular Manufacturing System Considering Order Rejection, Case Study: Stone Paper Factory. Engineering Management and Soft Computing, 10(2), 204-222.
Jafari, M., & Akbari, A. H. (2025). Efficient Algorithms for Dynamic Cellular Manufacturing Systems by Considering Blockchain-Enabled (Case Study: Stone Paper Factory). Journal of Advanced Manufacturing Systems.
ارسال نظر در مورد این مقاله
CAPTCHA Image
مدیریت مهندسی و رایانش نرم

مقالات آماده انتشار، پذیرفته شده
انتشار آنلاین از تاریخ 13 مرداد 1404

فایل ها

هم رسانی

ارجاع به این مقاله

آمار