Abstract
Human activity recognition has gained interest from the research community due to the advancements in sensor technology and the improved machine learning algorithm. Wearable sensors have become more ubiquitous, and most of the wearable sensor data contain rich temporal structural information that describes the distinct underlying patterns and relationships of various activity types. The nature of those activities is typically sequential, with each subsequent activity window being the result of the preceding activity window. However, the state-of-the-art methods usually model the temporal characteristic of the sensor data and ignore the relationship of the sliding window. This research proposes a novel deep temporal Conv-LSTM architecture to enhance activity recognition performance by utilizing both temporal characteristics from sensor data and the relationship of sliding windows. The proposed architecture is evaluated based on the dataset consisting of transition activities—Smartphone-Based Recognition of Human Activities and Postural Transitions dataset. The proposed hybrid architecture with parallel features learning pipelines has demonstrated the ability to model the temporal relationship of the activity windows where the transition of activities is captured accurately. Besides that, the size of sliding windows is studied, and it has shown that the selection of window size is affecting the accuracy of the activity recognition. The proposed deep temporal Conv-LSTM architecture can achieve an accuracy score of 0.916, which outperformed the state-of-the-art accuracy.
Similar content being viewed by others
Data Availability
Not applicable.
Code Availability
Not applicable/
References
Abidine BM, Fergani L, Fergani B, Oussalah M (2018) The joint use of sequence features combination and modified weighted SVM for improving daily activity recognition. Pattern Anal Appl 21:119–138. https://doi.org/10.1007/s10044-016-0570-y
Tian Y, Zhang J, Wang J et al (2020) Robust human activity recognition using single accelerometer via wavelet energy spectrum features and ensemble feature selection. Syst Sci Control Eng 8:83–96. https://doi.org/10.1080/21642583.2020.1723142
Vanrell SR, Milone DH, Rufiner HL et al (2018) Assessment of homomorphic analysis for human activity recognition from acceleration signals. IEEE J Biomed Health Inform 22:1001–1010. https://doi.org/10.1109/JBHI.2017.2722870
Ertuǧrul ÖF, Kaya Y (2017) Determining the optimal number of body-worn sensors for human activity recognition. Soft Comput 21:5053–5060. https://doi.org/10.1007/s00500-016-2100-7
Kanjilal R, Uysal I (2021) The future of human activity recognition: deep learning or feature engineering? Neural Process Lett 53:561–579. https://doi.org/10.1007/s11063-020-10400-x
Wang J, Chen Y, Hao S et al (2019) Deep learning for sensor-based activity recognition: a survey. Pattern Recognit Lett 119:3–11. https://doi.org/10.1016/j.patrec.2018.02.010
Xu W, Pang Y, Yang Y, Liu Y (2018) Human activity recognition based on convolutional neural network. In: 2018 24th International Conference on Pattern Recognition (ICPR), pp 165–170
Bevilacqua A, MacDonald K, Rangarej A et al (2019) Human activity recognition with convolutional neural networks. In: Brefeld U, Curry E, Daly E et al (eds) Machine learning and knowledge discovery in databases. Springer, Cham, pp 541–552
Lawal IA, Bano S (2019) Deep human activity recognition using wearable sensors. In: Proceedings of the 12th ACM International Conference on PErvasive Technologies Related to Assistive Environments. Association for Computing Machinery, New York, pp 45–48
Gil-Martín M, San-Segundo R, Fernández-Martínez F, Ferreiros-López J (2021) Time analysis in human activity recognition. Neural Process Lett 53:4507–4525. https://doi.org/10.1007/s11063-021-10611-w
Zhu R, Xiao Z, Cheng M et al (2018) Deep ensemble learning for human activity recognition using smartphone. In: 2018 IEEE 23rd International Conference on Digital Signal Processing (DSP), pp 1–5
Zehra N, Azeem SH, Farhan M (2021) Human activity recognition through ensemble learning of multiple convolutional neural networks. In: 2021 55th annual Conference on Information Sciences and Systems (CISS), pp 1–5
Sikder N, Chowdhury MdS, Arif ASM, Nahid A-A (2019) Human activity recognition using multichannel convolutional neural network. In: 2019 5th International Conference on Advances in Electrical Engineering (ICAEE), pp 560–565
Zhang H, Xiao Z, Wang J et al (2020) A novel IoT-perceptive Human Activity Recognition (HAR) approach using multihead convolutional attention. IEEE Internet Things J 7:1072–1080. https://doi.org/10.1109/JIOT.2019.2949715
Chen Y, Zhong K, Zhang J et al (2016) LSTM networks for mobile human activity recognition. Atlantis Press, pp 50–53
Zebin T, Sperrin M, Peek N, Casson AJ (2018) Human activity recognition from inertial sensor time-series using batch normalized deep LSTM recurrent networks. In: 2018 40th annual international conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp 1–4
Guan Y, Plötz T (2017) Ensembles of deep LSTM learners for activity recognition using wearables. Proc ACM Interact Mob Wearable Ubiquitous Technol 1:1–28. https://doi.org/10.1145/3090076
Li S, Li C, Li W et al (2018) Smartphone-sensors based activity recognition using IndRNN. In: Proceedings of the 2018 ACM International Joint Conference and 2018 International Symposium on Pervasive and Ubiquitous Computing and Wearable Computers. Association for Computing Machinery, New York, pp 1541–1547
Mahmud T, Akash SS, Fattah SA et al (2020) Human activity recognition from multi-modal wearable sensor data using deep multi-stage LSTM architecture based on temporal feature aggregation. In: 2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), pp 249–252
Ordóñez FJ, Roggen D (2016) Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors. https://doi.org/10.3390/s16010115
Mekruksavanich S, Jitpattanakul A (2020) Smartwatch-based human activity recognition using hybrid LSTM network. In: 2020 IEEE SENSORS, pp 1–4
Mutegeki R, Han DS (2020) A CNN-LSTM approach to human activity recognition. In: 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), pp 362–366
Li Z, Liu Y, Guo X, Zhang J (2020) Multi-convLSTM neural network for sensor-based human activity recognition. J Phys Conf Ser 1682:012062. https://doi.org/10.1088/1742-6596/1682/1/012062
Wang H, Zhao J, Li J et al (2020) Wearable sensor-based human activity recognition using hybrid deep learning techniques. Secur Commun Netw 2020:2132138. https://doi.org/10.1155/2020/2132138
Singh SP, Sharma MK, Lay-Ekuakille A et al (2021) Deep ConvLSTM with self-attention for human activity decoding using wearable sensors. IEEE Sens J 21:8575–8582. https://doi.org/10.1109/JSEN.2020.3045135
Abdel-Basset M, Hawash H, Chakrabortty RK et al (2021) ST-DeepHAR: deep learning model for human activity recognition in IoHT applications. IEEE Internet Things J 8:4969–4979. https://doi.org/10.1109/JIOT.2020.3033430
Xia K, Huang J, Wang H (2020) LSTM-CNN architecture for human activity recognition. IEEE Access 8:56855–56866. https://doi.org/10.1109/ACCESS.2020.2982225
Nafea O, Abdul W, Muhammad G, Alsulaiman M (2021) Sensor-based human activity recognition with spatio-temporal deep learning. Sensors. https://doi.org/10.3390/s21062141
Xiao Z, Xu X, Xing H et al (2021) A federated learning system with enhanced feature extraction for human activity recognition. Knowl -Based Syst 229:107338. https://doi.org/10.1016/j.knosys.2021.107338
Gao W, Zhang L, Teng Q et al (2021) DanHAR: dual attention network for multimodal human activity recognition using wearable sensors. Appl Soft Comput 111:107728. https://doi.org/10.1016/j.asoc.2021.107728
Reyes-Ortiz J-L, Oneto L, Sama A et al (2016) Transition-aware human activity recognition using smartphones. Neurocomputing 171:754–767
Janidarmian M, Roshan Fekr A, Radecka K, Zilic Z (2017) A Comprehensive analysis on wearable acceleration sensors in human activity recognition. Sensors. https://doi.org/10.3390/s17030529
Banos O, Galvez J-M, Damas M et al (2014) Window size impact in human activity recognition. Sensors 14:6474–6499. https://doi.org/10.3390/s140406474
Hogg RV, Tanis EA, Zimmerman DL (2010) Probability and statistical inference. Prentice Hall, Upper Saddle River
Funding
This work has been supported in part by the Ministry of Higher Education Malaysia for Fundamental Research Grant Scheme with Project Code: FRGS/1/2019/ICT02/USM/02/1.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Ethics Approval
Not applicable.
Consent to Participate
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Appendix
Window size: 80
A1 | A2 | A3 | A4 | A5 | A6 | A7 | A8 | A9 | A10 | A11 | A12 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
A1 | 890 | 23 | 1 | 0 | 0 | 0 | 4 | 2 | 0 | 0 | 1 | 2 |
A2 | 31 | 859 | 2 | 1 | 2 | 0 | 3 | 6 | 0 | 0 | 2 | 3 |
A3 | 4 | 38 | 818 | 0 | 0 | 0 | 2 | 0 | 0 | 0 | 1 | 1 |
A4 | 0 | 0 | 0 | 918 | 148 | 0 | 2 | 3 | 0 | 3 | 1 | 0 |
A5 | 4 | 1 | 0 | 150 | 964 | 2 | 2 | 1 | 0 | 0 | 1 | 2 |
A6 | 0 | 0 | 0 | 0 | 0 | 1143 | 0 | 1 | 0 | 0 | 1 | 2 |
A7 | 0 | 0 | 0 | 13 | 14 | 0 | 40 | 1 | 1 | 3 | 5 | 1 |
A8 | 0 | 0 | 0 | 6 | 16 | 0 | 1 | 36 | 0 | 1 | 0 | 2 |
A9 | 0 | 0 | 0 | 5 | 3 | 17 | 0 | 2 | 37 | 1 | 31 | 0 |
A10 | 0 | 0 | 0 | 14 | 0 | 8 | 1 | 1 | 0 | 50 | 0 | 6 |
A11 | 0 | 0 | 0 | 6 | 2 | 16 | 0 | 0 | 18 | 2 | 42 | 0 |
A12 | 3 | 0 | 0 | 4 | 7 | 11 | 0 | 6 | 0 | 25 | 0 | 29 |
Window size: 100
A1 | A2 | A3 | A4 | A5 | A6 | A7 | A8 | A9 | A10 | A11 | A12 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
A1 | 702 | 34 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 1 |
A2 | 19 | 696 | 13 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 |
A3 | 3 | 5 | 676 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 |
A4 | 0 | 0 | 0 | 710 | 139 | 0 | 2 | 0 | 0 | 4 | 2 | 0 |
A5 | 1 | 1 | 0 | 93 | 800 | 1 | 0 | 1 | 0 | 1 | 4 | 1 |
A6 | 0 | 0 | 0 | 0 | 0 | 918 | 0 | 0 | 0 | 0 | 2 | 1 |
A7 | 0 | 1 | 0 | 6 | 10 | 0 | 22 | 1 | 1 | 4 | 15 | 1 |
A8 | 0 | 1 | 0 | 5 | 11 | 0 | 1 | 33 | 0 | 0 | 0 | 1 |
A9 | 0 | 0 | 0 | 3 | 1 | 9 | 0 | 2 | 36 | 0 | 24 | 1 |
A10 | 0 | 0 | 0 | 12 | 0 | 6 | 0 | 0 | 0 | 35 | 1 | 8 |
A11 | 0 | 1 | 0 | 4 | 1 | 13 | 0 | 1 | 17 | 2 | 30 | 0 |
A12 | 1 | 2 | 0 | 2 | 3 | 7 | 0 | 2 | 0 | 22 | 0 | 28 |
Window size: 120
A1 | A2 | A3 | A4 | A5 | A6 | A7 | A8 | A9 | A10 | A11 | A12 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
A1 | 600 | 2 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 10 |
A2 | 31 | 562 | 5 | 0 | 1 | 0 | 0 | 2 | 0 | 0 | 0 | 0 |
A3 | 3 | 11 | 567 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
A4 | 0 | 0 | 0 | 606 | 99 | 0 | 4 | 1 | 0 | 3 | 0 | 0 |
A5 | 0 | 1 | 0 | 43 | 697 | 0 | 4 | 3 | 0 | 0 | 3 | 1 |
A6 | 0 | 0 | 0 | 0 | 0 | 762 | 0 | 0 | 4 | 1 | 1 | 1 |
A7 | 0 | 0 | 0 | 7 | 12 | 0 | 26 | 1 | 0 | 1 | 5 | 0 |
A8 | 0 | 0 | 0 | 4 | 7 | 0 | 5 | 24 | 0 | 0 | 0 | 0 |
A9 | 0 | 0 | 0 | 5 | 3 | 11 | 0 | 0 | 24 | 0 | 21 | 0 |
A10 | 0 | 0 | 0 | 8 | 0 | 3 | 3 | 0 | 1 | 33 | 0 | 7 |
A11 | 0 | 0 | 0 | 4 | 2 | 6 | 4 | 0 | 13 | 0 | 28 | 0 |
A12 | 1 | 0 | 0 | 5 | 3 | 4 | 0 | 2 | 0 | 16 | 0 | 22 |
Window size: 140
A1 | A2 | A3 | A4 | A5 | A6 | A7 | A8 | A9 | A10 | A11 | A12 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
A1 | 499 | 21 | 3 | 0 | 0 | 0 | 2 | 1 | 0 | 0 | 0 | 4 |
A2 | 12 | 484 | 10 | 0 | 1 | 0 | 0 | 10 | 0 | 0 | 0 | 5 |
A3 | 0 | 5 | 481 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
A4 | 0 | 0 | 0 | 503 | 98 | 0 | 2 | 3 | 1 | 3 | 0 | 1 |
A5 | 0 | 0 | 0 | 77 | 552 | 0 | 4 | 8 | 0 | 1 | 3 | 0 |
A6 | 0 | 0 | 0 | 0 | 0 | 653 | 0 | 0 | 1 | 1 | 0 | 3 |
A7 | 0 | 0 | 0 | 5 | 12 | 0 | 25 | 0 | 0 | 0 | 1 | 0 |
A8 | 0 | 0 | 0 | 1 | 3 | 0 | 0 | 28 | 0 | 2 | 0 | 0 |
A9 | 0 | 0 | 0 | 3 | 0 | 5 | 1 | 2 | 21 | 0 | 23 | 0 |
A10 | 0 | 0 | 0 | 9 | 0 | 3 | 1 | 0 | 0 | 29 | 0 | 6 |
A11 | 0 | 0 | 0 | 3 | 1 | 7 | 4 | 0 | 8 | 0 | 28 | 0 |
A12 | 0 | 0 | 0 | 1 | 2 | 4 | 0 | 0 | 0 | 19 | 0 | 21 |
Window size: 128
A1 | A2 | A3 | A4 | A5 | A6 | A7 | A8 | A9 | A10 | A11 | A12 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
A1 | 525 | 49 | 4 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 |
A2 | 15 | 446 | 6 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |
A3 | 0 | 10 | 526 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
A4 | 0 | 0 | 0 | 594 | 68 | 0 | 1 | 4 | 0 | 1 | 0 | 0 |
A5 | 1 | 1 | 0 | 21 | 674 | 2 | 1 | 2 | 0 | 1 | 5 | 0 |
A6 | 0 | 0 | 0 | 0 | 0 | 713 | 0 | 0 | 0 | 1 | 1 | 4 |
A7 | 0 | 0 | 0 | 11 | 6 | 0 | 27 | 2 | 0 | 1 | 2 | 0 |
A8 | 1 | 0 | 0 | 2 | 5 | 0 | 0 | 32 | 0 | 0 | 0 | 0 |
A9 | 0 | 0 | 0 | 5 | 1 | 7 | 0 | 0 | 20 | 0 | 25 | 1 |
A10 | 0 | 0 | 0 | 9 | 2 | 2 | 0 | 0 | 0 | 28 | 0 | 7 |
A11 | 0 | 0 | 0 | 1 | 3 | 6 | 0 | 1 | 5 | 0 | 39 | 0 |
A12 | 1 | 0 | 0 | 2 | 2 | 2 | 0 | 2 | 0 | 17 | 0 | 25 |
Rights and permissions
About this article
Cite this article
Mohd Noor, M.H., Tan, S.Y. & Ab Wahab, M.N. Deep Temporal Conv-LSTM for Activity Recognition. Neural Process Lett 54, 4027–4049 (2022). https://doi.org/10.1007/s11063-022-10799-5
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11063-022-10799-5