Abstract
A reliable adaptive hybrid classifier (hAHC), which combines a posture-based adaptive signal segmentation algorithm with a multi-layer perceptron (MLP) classifier, together with a plurality voting approach, was proposed and evaluated in this study. The hAHC model was evaluated using a real-time posture recognition framework that sought to identify five behaviours (sitting, walking, standing, running, and lying) based on simulated crowd security scenarios. It was compared to a single MLP classifier (sMLP) and a static hybrid classifier (hSHC) from three perspectives (classification precision, recall and F1-score) that used the real-time dataset collected from unfamiliar subjects. Experimental results showed that the hAHC model improved the classification accuracy and robustness slightly more than the hSHC, and significantly more compared to the sMLP (hAHC 82%; hSHC 79%; sMLP 71%). Additionally, the hAHC approach displayed the real-time results as animated figures in an adaptive window, in contrast to the hSHC which used a fixed size-sliding temporal window that as our results demonstrated, was less suitable for presenting real-time results. The main research contribution from this study has been the development of an efficient software-only-based sensor calibration algorithm that can improve accelerometer precision, together with the design of a posture-based adaptive signal segmentation algorithm that cooperated with an adaptive hybrid classifier to improve the performance of real-time posture recognition.
Similar content being viewed by others
References
Liu T, Kong J, Jiang M et al (2019) Collaborative model with adaptive selection scheme for visual tracking. Int J Mach Learn Cyber 10:215–228
Cai X, Han G, Song X et al (2019) Gait symmetry measurement method based on a single camera. Int J Mach Learn Cyber 10:1399–1406
Khan G, Samyan S, Khan MUG et al (2020) A survey on analysis of human faces and facial expressions datasets. Int J Mach Learn Cyber 11:553–571
Cao Y, Xue F, Chi Y et al (2020) Effective spatio-temporal semantic trajectory generation for similar pattern group identification. Int J Mach Learn Cyber 11:287–300
Mabrouk AB, Zagrouba E (2018) Abnormal behavior recognition for intelligent video surveillance systems: a review. Expert Syst Appl. 91(1):480–491
Khokhlov I, Rezni, L, Cappos J, Bhaskar R (2018) Design of activity recognition systems with wearable sensors. IEEE Sensors Applications Symposium (SAS), pp. 1–6
Zhang S, McCullagh P, Zheng H, Nugent C (2017) Situation awareness inferred from posture transition and location: derived from smartphone and smart home sensors. IEEE Trans Hum Mach Syst 47(6):814–821
Freeston IL, Callaghan VL, Russel ND (1984). A portable navigation aid for the blind, in frontiers of engineering and computing in health care. IEEE Conference on Engineering in Medicine and Biology Society, Los Angeles, CA, pp. 247–249
Cornacchia M, Ozcan K, Zheng Y, Velipasalar S (2017) A survey on activity detection and classification using wearable sensors. IEEE Sens J 17(2):386–403
Cheok MJ, Omar Z, Jaward MH (2019) A review of hand gesture and sign language recognition techniques. Int J Mach Learn Cyber 10(9):131–153
Bonnet S, Bassompierre C, Godin C, Lesecq S, Barraud A (2009) Calibration methods for inertial and magnetic sensors. Sens Actuators A 156(2):302–311
Xue B, Zhang M, Browne WN, Yao X (2016) A survey on evolutionary computation approaches to feature selection. IEEE Trans Evol Comput 20(4):606–626
Vari A (1988), Digital processing of the EEG in epilepsy. Licentiate thesis. Tampere University of Technology, pp. 1–97
Azami H, Mohammadi K, Hassanpour H (2011) An improved signal segmentation method using genetic algorithm. Int J Comput Appl 29(8):5–9
Azami H, Mohammadi K, Bozorgtabar B (2012) An improved signal segmentation using moving average and Savitzky-Golay filter. J Signal Inf Process 3(01):39
Novosadová M, Rajmic P, Šorel M (2019) Orthogonality is superiority in piecewise-polynomial signal segmentation and denoising. EURASIP J Adv Signal Process 6(2019):1–22
Laguna JO, Olaya AG, Borrajo D (2011) A dynamic sliding window approach for activity recognition. In: International conference on user modeling, adaptation, and personalization, pp. 219–230
Gao L, Bourke AK, Nelson J (2014) Evaluation of accelerometer based multi-sensor versus single-sensor activity recognition systems. Med Eng Phys 36(6):779–785
Saini R, Kumar P, Kaur B et al (2019) Kinect sensor-based interaction monitoring system using the BLSTM neural network in healthcare. Int J Mach Learn Cyber 10(9):2529–2540
Hegde N, Bries M, Swibas T, Melanson E, Sazonov E (2017) Automatic recognition of activities of daily living utilizing insole based and wrist worn wearable sensors. IEEE J Biomed Health Inf
Marcon M, Paracchini MBM, Tubaro S (2019) A framework for interpreting, modeling and recognizing human body gestures through 3D eigenpostures. Int J Mach Learn Cyber 10:1205–1226
Naveed H, Khan G, Khan AU, Siddiqi A, Khan MUG (2019) Human activity recognition using mixture of heterogeneous features and sequential minimal optimization. Int J Mach Learn Cyber 10:2329–2340
https://openrtls.com/shop. Accessed May 2018
Zhang S, Monekosso D, Remagnino P (2018) Data pre-processing and model selection strategies for human posture recognition. IEEE international symposium on communication systems, networks & digital signal processing, pp. 1–6
Rohac J, Sipos M, Simanek J (2015) Calibration of low-cost triaxial inertial sensors. IEEE Instrum Meas Mag 18(6):32–38
Christian JA, Bittner DE, West Virginia University (2018) Apparatus for three-axis IMU calibration with a single-axis rate table. U.S. Patent 9,970,781
Hall JJ, Williams RL (2000) Case study: inertial measurement unit calibration platform. J Robot Syst 17(11):623–632
Harris C, Mahajan S, Leslie P, Spani C et al (2018) Wireless inertial measurement system provides accurate dynamic angle change and angular data for body mount applications. CMBES Proc 33(1)
Dunne RA, Campbell NA (1997) On the pairing of the softmax activation and cross-entropy penalty functions and the derivation of the softmax activation function. In Proceedings of 8th Aust. Conference on the Neural Networks, Melbourne, vol 181, p. 185
Truong C, Oudre L, Vayatis N (2018) Ruptures: change point detection in Python. arXiv preprint arXiv:1801.00826
Yang J, Nguyen MN, San PP, Li XL, Krishnaswamy S (2015) Deep convolutional neural networks on multichannel time series for human activity recognition. International Joint Conference on Artificial Intelligence
Zebin T, Scully PJ, Ozanyan KB (2016) Human activity recognition with inertial sensors using a deep learning approach. In 2016 IEEE SENSORS, pp. 1–3
Luca O (2020) Model selection and error estimation in a nutshell. Springer International Publishing, Berlin
Acknowledgement
The authors are pleased to acknowledge funding from European Union, H2020 Grant No. 732350 and the IoT application technology innovation centre of Hebei province, China, Grant No. SG20182058. We wish to convey a special thanks to Prof. Dorothy Monekosso, the H2020 grant holder at Leeds Beckett University, for co-opting the lead author into this project. We also want to express our gratitude to various staff members at Leeds Beckett University for their kind assistance; and IT Services staff for the loan of camcorders used in the video recording of the experiments. Finally, we should not forget the anonymised people who provided the experimental data, thank you all for supporting this research.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Zhang, S., Callaghan, V. Real-time human posture recognition using an adaptive hybrid classifier. Int. J. Mach. Learn. & Cyber. 12, 489–499 (2021). https://doi.org/10.1007/s13042-020-01182-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13042-020-01182-8