Biometrics gait system based on motion sensors embedded in a mobile phone: a case study for a two-day training set

• Autorzy: Sawicki Aleksander , Saeed Khalid

Identyfikatory

DOI

10.24425/ijet.2025.153603

• Języki publikacji: EN

Abstrakty

EN

This paper presents the results of a study on developing a gait biometrics system based on motion sensors (an accelerometer and gyroscope), embedded in a smartphone. The experiments were conducted using a publicly available 13-person data corpus, with subjects participating in three data collection sessions. The study used CNN, CNN with attention and Multi-Input CNN neural networks. The training scenario from the first day resulted in an accuracy of 0.66 F1 score, 0.71 F1 score for training with the samples from the second day and 0.90 F1 score in the combined sets. It has been shown that it is more profitable to combine historical data than to update it with newer samples. Enriching the training set with a set of 30% synthetic samples produced by the LSTM-MDN generative models allowed to increase to accuracy to 0.94 F1-score. It was shown that synthetic samples can improve the generalization properties of the CNN network.

Słowa kluczowe

EN

CNN; biometrics; IMU; deep learning

• Wydawca: Polish Academy of Sciences, Committee of Electronics and Telecommunication
• Czasopismo: International Journal of Electronics and Telecommunications
• Rocznik: 2025
• Tom: Vol. 71, No. 2
• Strony: 545--553
• Opis fizyczny: Bibliogr. 20 poz., tab., rys.

Twórcy

autor

Sawicki Aleksander

• Białystok University of Technology, Poland

• https://orcid.org/0000-0002-8662-4484

autor

Saeed Khalid

• Białystok University of Technology, Poland

• https://orcid.org/0000-0002-7741-7045

Bibliografia

• [1] Q. Zou, Y. Wang, Q. Wang, Y. Zhao, and Q. Li, ‘Deep Learning-Based Gait Recognition Using Smartphones in the Wild’, arXiv [cs.LG]. 2020. https://doi.org/10.48550/arXiv.1811.00338
• [2] S. Sprager and M. B. Juric, ‘Inertial Sensor-Based Gait Recognition: A Review’, Sensors, vol. 15, no. 9, pp. 22089-22127, 2015. https://doi.org/10.3390/s150922089
• [3] G. Giorgi, F. Martinelli, A. Saracino, and M. Alishahi, ‘Try Walking in My Shoes, if You Can: Accurate Gait Recognition Through Deep Learning’, 09 2017, pp. 384-395. https://doi.org/10.1007/978-3-319-66284-8_
• [4] C. Wan, L. Wang, and V. V. Phoha, ‘A survey on gait recognition’, ACM Computing Surveys, vol. 51, no. 5, Aug. 2018. https://doi.org/10.1145/3230633
• [5] A. Ajit, N. K. Banerjee, and S. Banerjee, ‘Combining Pairwise Feature Matches from Device Trajectories for Biometric Authentication in Virtual Reality Environments’, in 2019 IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR), 2019, pp. 9-97. https://doi.org/10.1109/AIVR46125.2019.00012
• [6] J. E. Boyd and J. J. Little, ‘Biometric Gait Recognition’, in Advanced Studies in Biometrics: Summer School on Biometrics, Alghero, Italy, June 2-6, 2003. Revised Selected Lectures and Papers, M. Tistarelli, J. Bigun, and E. Grosso, Eds. Berlin, Heidelberg: Springer Berlin Heidelberg, 2005, pp. 19-42. https://doi.org/10.1007/11493648_2
• [7] R. Plucińska, K. Jędrzejewski, U. Malinowska, and J. Rogala, ‘Influence of Feature Scaling and Number of Training Sessions on EEG Spectral-based Person Verification with Artificial Neural Networks’, in 2023 Signal Processing Symposium (SPSympo), 2023, pp. 139-143. https://doi.org/10.23919/SPSympo57300.2023.10302695
• [8] N. Al-Naffakh, N. Clarke, and F. Li, ‘Continuous User Authentication Using Smartwatch Motion Sensor Data’, in Trust Management XII, 2018, pp. 15-28. https://doi.org/10.1007/978-3-319-95276-5_2
• [9] D. S. Matovski, M. S. Nixon, S. Mahmoodi, and J. N. Carter, ‘The Effect of Time on Gait Recognition Performance’, IEEE Transactions on Information Forensics and Security, vol. 7, no. 2, pp. 543-552, 2012. https://doi.org/10.1109/TIFS.2011.2176118
• [10] S. Lee, S. Lee, E. Park, J. Lee, and I. Y. Kim, ‘Gait-Based Continuous Authentication Using a Novel Sensor Compensation Algorithm and Geometric Features Extracted From Wearable Sensors’, IEEE Access, vol. 10, pp. 120122-120135, 2022. https://doi.org/10.1109/ACCESS.2022.3221813
• [11] A. Sawicki and K. Saeed, 'Smartphone-Based Biometric System Involving Multiple Data Acquisition Sessions', System Dependability - Theory and Applications. DepCoS-RELCOMEX 2024. Lecture Notes in Networks and Systems, vol 1026, 2024, pp 252-260. https://doi.org/10.1007/978-3-031-61857-4_25
• [12] M. Gadaleta and M. Rossi, ‘IDNet: Smartphone-based gait recognition with convolutional neural networks’, Pattern Recognition, vol. 74, pp. 25-37, 2018. https://doi.org/10.1016/j.patcog.2017.09.005
• [13] M. W. Whittle, ‘Chapter 2 - Normal gait’, in Gait Analysis (Fourth Edition), Fourth Edition., M. W. Whittle, Ed. Edinburgh: Butterworth-Heinemann, 2007, pp. 47-100. https://doi.org/10.1016/B978-0-7506-8883-3.X5001-6
• [14] A. Desai, C. Freeman, Z. Wang, and I. Beaver, ‘TimeVAE: A Variational Auto-Encoder for Multivariate Time Series Generation’, arXiv [cs.LG]. 2021. https://doi.org/10.48550/arXiv.2111.08095
• [15] C. Chadebec, E. Thibeau-Sutre, N. Burgos, and S. Allassonnière, ‘Data Augmentation in High Dimensional Low Sample Size Setting Using a Geometry-Based Variational Autoencoder’, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 3, pp. 2879-2896, Mar. 2023. https://doi.org/10.48550/arXiv.2105.00026
• [16] A. Sawicki and D. Grabowski 'Application of Mixture Density Network for Sample Generation in Behavioral Biometrics', Computer Information Systems and Industrial Management. CISIM 2024. Lecture Notes in Computer Science, vol 14902 (2024), pp 30-43. https://doi.org/10.1007/978-3-031-71115-2_3
• [17] H. Huang, P. Zhou, Y. Li, and F. Sun, ‘A Lightweight Attention-Based CNN Model for Efficient Gait Recognition with Wearable IMU Sensors’, Sensors, vol. 21, no. 8, 2021. https://doi.org/10.3390/s21082866
• [18] R. Delgado-Escaño, F. M. Castro, J. R. Cózar, M. J. Marín-Jiménez, and N. Guil, ‘An End-to-End Multi-Task and Fusion CNN for Inertial-Based Gait Recognition’, IEEE Access, vol. 7, pp. 1897-1908, 2019. https://doi.org/10.1109/ACCESS.2018.2886899
• [19] Li, X., Luo, J., Younes, R.: ActivityGAN: generative adversarial networks for data augmentation in sensor-based human activity recognition. In Adjunct Proc. of the ACM Int. Joint Conf. on Pervasive and Ubiquitous Computing and Proc. of the ACM Int. Sym. on Wearable Computers Association for Computing Machinery, 249-254 (2020), https://doi.org/10.1145/3410530.3414367
• [20] D. Carneros-Prado, C. C. Dobrescu, L. Cabañero, L. Villa, Y. V. Altamirano-Flores et al.‘ Synthetic 3D full-body skeletal motion from 2D paths using RNN with LSTM cells and linear networks’, Computers in Biology and Medicine, Volume 180, 2024, 108943, ISSN 0010-4825 https://doi.org/10.1016/j.compbiomed.2024.108943.

Uwagi

1. Opracowanie rekordu ze środków MNiSW, umowa nr POPUL/SP/0154/2024/02 w ramach programu "Społeczna odpowiedzialność nauki II" - moduł: Popularyzacja nauki (2025).

2. This work was supported by grant 2021/41/N/ST6/02505 and funded with resources for research by National Science Centre, Poland. For the purpose of Open Access, the author has applied a CC-BY public copyright license to any Author Accepted Manuscript (AAM) version arising from this submission.