Walking-posture Classification from Single-acceleration-sensor Data using Deep Learning and its Evaluation
Abstract
We described a walking-posture classification method from a single accelerator attached to a human waist using a deep learning technique. We considered deep learning architectures for a single accelerator based on previous human activity recognition studies and investigated the classification accuracy of the proposed method using the walking-posture dataset. The results demonstrate that a deep learning approach to walking-posture classification using a single accelerator is more useful than the conventional SVM approach. Additionally, we also confirmed that a hybrid network architecture with three convolutional neural layers, two pooling layers between the convolutional layers, and a long short-term memory layer achieved the best accuracy of 0.9803 compared to other network architectures. We also confirmed the deep learning approach yielded more correct classification with less periods of each sample for each walking-posture category in spite of the difficulty to detect the classification by the SVM approach.
References
Almaslukh B., AlMuhtadi J., Artoli A. An Effective Deep Autoencoder Approach for Online Smartphone-based Human Activity Recognition. Int. J. Comput. Sci. Netw. Secur. 17:160, 2017.
Chan, H., Zheng, H., Wang, H., Sterritt R., Newell, D. Smart Mobile Phone based Gait Assessment of Patients with Low Back Pain, Procs. 9th International Conference on Natural Computation, pp.1062-1066, 2013.
Chung, J., Caglar G., Cho, K. and Bengio, Y. Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. ArXiv abs/1412.3555, 2014.
Gadaleta, M., Rossi, M. IDNet: Smartphone-based Gait Recognition with Convolutional Neural Networks, Pattern Recognition, Vol.74, pp.25–37, 2018.
Karvekar, S. B. Smartphone-based Hauman Fatigue Detection in an Industrial Environment Using Gait Analysis. Thesis. Rochester Institute of Technology, 2019.
Hammerla, N. Y., Fisher, J. M., Andras, P., Rochester, L., Walker, R., Plotz., T.PD Disease State Assessment in Naturalistic Environments Using Deep Learning. Procs. the 29th AAAI Conference on Artificial Intelligence. AAAI Press, pp.1742-1748, 2015.
Hashiguchi, H., Hori, K., Hirobe, Y., Sawada, H. Inaba, A., Orimo, S., Miyake, Y. Early Detection System of Parkinson’s Disease Based on the Walking Trajectory Data. the 28th SICE Symposium on Decentralized Autonomous Systems, pp.155-158, 2016 (in Japanese).
Hochreiter, S., Schmidhuber, J. Long short-term memory. Neural Computation 9 (8),pp.1735–1780, 1997.
Jordao, A., Nazare Jr, A. C., Sena, J., Schwartz W. R. Human Activity Recognition Based on Wearable Sensor Data: A Standardization of the State-of-the-art, arXiv preprint arXiv:1806.05226, 2018.
Kirsten G. N. Observational Gait Instructor Group, IGAKU-SHOIN Ltd., 2005.
Lotfi, A., Nguyen, M., Langensiepen, C. Human Gait Classification Using a Tri-axial Accelerometer. Procs. the 8th ACM International Conference on PErvasive Technologies Related to Assistive Environments, Article 36, pp.1-4, 2015.
Ordonez, F., Roggen, D. Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition. Sensors 16(1), pp.1533–1540, 2016.
Plotz, T., Hammerla, N. Y., Olivier, P. Feature learning for activity recognition in ubiquitous computing. Procs. the 22nd IJCAI, Vol.2. AAAI Press, pp.1729-1734, 2011.
Nickel, C., Busch, C., Rangarajan, S., Mobius, M. Using Hidden Markov Models for accelerometer-based biometric gait recognition, IEEE 7th International Colloquium on Signal Processing and its Applications, Penang, pp.58-63, 2011.
Yao, S., Hu, S., Zhao, Y., Zhang, A., Abdelzaher, T. Deepsense: A unified Deep Learning framework for time-series mobile sensing data processing. Procs. the 26th International Conference on World Wide Web, pp.351–360, 2017.
Yang, J. B., Nguyen, M. N., San, P. P., Li, X. L., Krishnaswamy, S. Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition. Procs. the 24th International Conference on Artificial Intelligence. AAAI Press, pp.3995-40010, 2015.
Wang, J., Chen, Y., Hao, S., Peng, X., Hu, L. Deep Learning for Sensor-based Activity Recognition: A Survey. Computer Vision. Pattern Recognition Letters. pp.3-11, 2017.
