Implementation of Deep Learning-Based Motion Classification System for IoT Device Control in Ultrasonic Sound Environments 


Vol. 42,  No. 9, pp. 1796-1805, Sep.  2017
10.7840/kics.2017.42.9.1796


PDF Full-Text
  Abstract

There are lots of IoT devices control methods using speech recognition or gesture recognition. In this paper, we propose a new gesture recognition method to control IoT devices using deep-learning technique. In the proposed method, we generate ultrasonic sound signals between speaker and microphone. During the generation of a sinusoidal wave signal, we can obtain a 2-dimension frequency-time domain data which captures doppler effect in accordance to specific gesture. In our system, we classify five different gestures using convolutional neural network (CNN). The classification index that is obtained by CNN is transmitted to the remote side raspberry pi device using MQTT protocol. The raspberry pi can control the IoT devices by the received data. Simulation results show that we can obtain the high classification accuracy for gestures and also can effectively control the IoT devices using gestures.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Yang, W. Song, I. Choi, S. Yoo, "Implementation of Deep Learning-Based Motion Classification System for IoT Device Control in Ultrasonic Sound Environments," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 9, pp. 1796-1805, 2017. DOI: 10.7840/kics.2017.42.9.1796.

[ACM Style]

Su-Myung Yang, Won-Jae Song, Ik-Soo Choi, and Sang-Jo Yoo. 2017. Implementation of Deep Learning-Based Motion Classification System for IoT Device Control in Ultrasonic Sound Environments. The Journal of Korean Institute of Communications and Information Sciences, 42, 9, (2017), 1796-1805. DOI: 10.7840/kics.2017.42.9.1796.

[KICS Style]

Su-Myung Yang, Won-Jae Song, Ik-Soo Choi, Sang-Jo Yoo, "Implementation of Deep Learning-Based Motion Classification System for IoT Device Control in Ultrasonic Sound Environments," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 9, pp. 1796-1805, 9. 2017. (https://doi.org/10.7840/kics.2017.42.9.1796)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880