Abstract
The traditional human behavior recognition technology mainly includes sign point action recognition technology and recognition technology based on motion sensor parameters. The former’s error is very large and the latter’s recognition speed is slow as well as the accuracy is very low. In this paper, a method of recognition of depth learning behavior based on convolutional neural network is proposed to identify different behaviors (jogging, walking, running, upstairs, downstairs, sitting) with different location of mobile phone (arm, waist, pocket, wrist), collecting a large amount of data by using the built-in sensor of the mobile phone. The data are standardized, normalized and window segmentation, and then the data are divided into testing set and training set. Establish a convolutional neural network learning model to extract local feature structure and combine supervised learning mode, use training set for training, and then use testing set for classifying and evaluating. Through experiments, in the common movements and different placement positions, the accuracy is more than 94%, and the effect of high speed and high accuracy to identify human behavior is achieved.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Wei, C.: Classification and recognition of human behavior in video image sequences. Yanshan University (2012)
Mi, X., Li, X.: Study on the automatic monitoring method of elderly living alone based on internet of things intelligence. Comput. Simul. 31(02), 378–381 (2014)
Liu, N.: Recognition and research of fall behavior based on multi-feature fusion. Hebei University of Technology (2015)
Xue, F.: Research and application of behavior detection based on smart phone sensor. Southwest University (2017)
Zhang, Y.: Human behavior recognition based on smart phone sensor data [OL], 21 February 2017. http://www.infoq.com/cn/articles/human-behavior-recognition-based-on-smart-phone-sensor-data?utm_source=tuicool&utm_medium=referral
Ponce, H., Martínez-Villaseñor, M.L., Miralles-Pechuán, L.: A novel wearable sensor-based human activity recognition approach using artificial hydrocarbon networks. Sensors (Basel, Switzerland) 16(7), 1033 (2016)
Liu, L., Peng, Y., Liu, M., Huang, Z.: Sensor-based human activity recognition system with a multilayered model using time series shapelets. Knowl. Based Syst. 90, 138–152 (2015)
Liu, C., Zhang, L., Liu, Z., et al.: Lasagna: towards deep hierarchical understanding and searching over mobile sensing data. In: International Conference on Mobile Computing and Networking, pp. 334–347. ACM (2016)
Wu, D., Wang, Z., Chen, Y., Zhao, H.: Mixed-kernel based weighted extreme learning machine for inertial sensor based human activity recognition with imbalanced dataset. Neurocomputing 190, 35–49 (2016)
Zhang, X.: Deep learning algorithm and application based on convolutional neural network. Xidian University (2015)
Deng, J., Zhang, X.: Deep learning optimization method based on automatic encoder combination. Comput. Appl. 36(03), 697–702 (2016)
Ji, S., Xu, W., Yang, M., Yu, K.: 3D convolutional neural networks for human action recognition. IEEE Trans. Pattern Anal. Mach. Intell. 35(1), 221–231 (2013)
Zhang, H.Y.: Simulation line design and its FPGA realization based on bp neural network. J. Electron. Inf. Technol. 29(5), 1267–1270 (2007)
Luo, P., Li, H.: Quantum neural network algorithm based on Tanh multilayer function and its application. Comput. Digit. Eng. 40(1), 4–6 (2012)
Wang, D., Liu, S.: A new modified activation unit LogReLU using log function. J. Jilin Univ. (Sci. Ed.) 55(3), 617–622 (2017)
Liu, Q., Tang, X., Zhang, N.: Structural optimization convolutional neural network based on unsupervised pre-training. Eng. Sci. Technol. 2017(s2): 210–215 (2017)
Boureau, Y., Bach, F., Lecun, Y., et al.: Learning mid-level features for recognition. In: Computer Vision and Pattern Recognition, pp. 2559–2566. IEEE (2010)
Charalampous, K., Kostavelis, I., Gasteratos, A.: Robot navigation in large-scale social maps: an action recognition approach. Expert Syst. Appl. 66(C), 261–273 (2016)
Wang, G., Duan, M., Niu, C.: A stochastic gradient descent algorithm based on convolutional neural network. Comput. Eng. Des. 39(02): 441–445+462 (2018)
Liu, Q., Guo, J., Pu, H., Yan, Z.: Four-rotor UAV attitude estimation system based on gradient descent method [J/OL]. Electrooptics and Control, pp. 1–7, 11 April 2018. http://kns.cnki.net/kcms/detail/41.1227.TN.20171213.0935.020.html
Chen, S.: Research on deep learning neural network in speech recognition. South China University of Technology (2013)
Wang, Y.: Design and implementation of image recognition and text recommendation system based on deep learning. Beijing Jiaotong University (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Ma, P., Zou, T., Wang, Y. (2019). Research on Human Behavior Recognition Based on Convolutional Neural Network. In: Shen, S., Qian, K., Yu, S., Wang, W. (eds) Wireless Sensor Networks. CWSN 2018. Communications in Computer and Information Science, vol 984. Springer, Singapore. https://doi.org/10.1007/978-981-13-6834-9_12
Download citation
DOI: https://doi.org/10.1007/978-981-13-6834-9_12
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-13-6833-2
Online ISBN: 978-981-13-6834-9
eBook Packages: Computer ScienceComputer Science (R0)