Adaptive multiple classifiers fusion for inertial sensor based human activity recognition
- 216 Downloads
Aiming at the poor accuracy of single classifier in recognizing daily activities based on single accelerometer, this paper presents a method of daily activity recognition based on ensemble learning and full information matrix based fusion weight. Firstly, features from three attributes are extracted from the acceleration signals respectively. The three kinds of features can well describe the information of the activity, and they are relatively independent, which can reduce the interference caused by information redundancy in the process of fusion. Then three base classifiers of support vector machines are constructed based on three kinds of features respectively. Secondly, Euclidean distance between the test sample and every training sample for each type of feature vector is calculated to find out the k nearest neighbors of the test sample from the training set by the K-nearest neighbour method. The cluster analysis is used to compute the similarity between every neighbor and the test sample. Then, a proper threshold is utilized to remove the invalid neighbor whose similarity is less than the threshold. According to the effective neighbor, the full information matrix is constructed to calculate the accuracy. The weight of every single classifier is set dynamically according to the accuracy. Experiments showed that our proposed method get the best average recognition accuracy of 94.79% among several other weight functions when using majority voting method, besides, the time cost is also appealing.
KeywordsHuman activity recognition Multiple classifiers fusion Ensemble learning system Full information matrix Wearable triaxial accelerometer Adaptive weighted
This work was supported by the National Key Technology Research and Development Program of the Ministry of Science and Technology of China No. 2015BAI06B03.
- 3.King, R.C., Atallah, L., Wong, C., Miskelly, F., Yang, G.Z.: Elderly risk assessment of fall with BSN. In: Proceedings of 2010 International Conference on Body Sensor Networks, pp. 30–35 (2010)Google Scholar
- 6.Zhu, C., Sheng, W.: Human daily activity recognition in robot-assisted living using multi-sensor fusion. In: IEEE International Conference on Robotics and Automation, pp. 2154–2159 (2009)Google Scholar
- 14.Carós, J.S., Chételat, O., Celka, P., Dasen, S., CmÃral, J.: Very low complexity algorithm for ambulatory activity classification. In: 3rd European Medical and Biological Conference EMBEC, pp. 16–20 (2005)Google Scholar
- 18.Feng, Z.T., Mo, L.F., Li, M.: A Random Forest-based ensemble method for activity recognition. In: 37th Annual International Conference of the IEEE Engineering in Medicine and and Biology Society (EMBC), pp. 5074–5077 (2015)Google Scholar
- 19.Lester, J., Choudhury, T., Borriello, G.: A practical approach to recognizing physical activities. In: 2006 Pervasive Computing. Lecture Notes in Computer Science, vol. 3968, pp. 1–16. Springer, BerlinGoogle Scholar
- 20.Lester, J., Choudhury, T., Kern, N., Borriello, G., Hannaford, B.: A hybrid discriminative/generative approach for modeling human activities. In: 2005 19th International Joint Conference on Artificial Intelligence (Edinburgh), pp. 766–772Google Scholar
- 22.Daghistani, T., Alshammari, R.: Improving accelerometer-based activity recognition by using ensemble of classifiers. Int. J. Adv. Comput. Sci. Appl. 7(5), 128–133 (2016)Google Scholar
- 26.Maghsoudi, Y., Alimohammadi, A., Valadan Zoej, M.J., Mojaradi, B.: Weighted combination of multiple classifiers for the classification of hyperspectral images using a genetic algorithm, presented at the SPRS Commiss. VII, Mid-term Symp., Remote Sens.: Pixels Processes, Enschede, The Netherlands (2006)Google Scholar
- 27.Li, Z.P., Fan, Q.J., Chang, L.M., Yang, X.H.: Improved wavelet threshold denoising method for MEMS Gyroscope. In: 11th IEEE International Conference on Control & Automation (ICCA), pp. 530–534 (2014)Google Scholar
- 31.Sun, J., Wang, C., Du, Q.: Analog circuit fault diagnosis based on wavelet packet energy spectrum and NPE. Chin. J. Sci. Instrum. 34, 2021–2027 (2013)Google Scholar