Implementation of Fall Detection Based on CNN-LSTM 


Vol. 47,  No. 2, pp. 340-347, Feb.  2022
10.7840/kics.2022.47.2.340


PDF
  Abstract

In this paper, we implement a fall detection system through combining CNN(convolutional neural network) and LSTM(long short term memory) that are kinds of image data processing and sequence data processing, respectively. A dataset of three common human postures(standing posture, sitting posture, and lying posture) was collected using a non-contact infrared thermal array sensor. The output is obtained by employing a CNN-LSTM combination method, in which the sample data is first fed into the CNN, and the output of the CNN is then utilized as an input for the LSTM network. In this paper, the accuracy was compared and analyzed after implementing fall detection systems in two deep learning models,(CNN-LSTM, ResNetCNN-LSTM), through video datasets collected using non-contact infrared thermal array sensor. Future research directions are proposed based on the accuracy of the experimental data.

  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]

Y. Lee, J. Park, S. Shin, "Implementation of Fall Detection Based on CNN-LSTM," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 2, pp. 340-347, 2022. DOI: 10.7840/kics.2022.47.2.340.

[ACM Style]

Yeong-wook Lee, Jae-han Park, and Soo-young Shin. 2022. Implementation of Fall Detection Based on CNN-LSTM. The Journal of Korean Institute of Communications and Information Sciences, 47, 2, (2022), 340-347. DOI: 10.7840/kics.2022.47.2.340.

[KICS Style]

Yeong-wook Lee, Jae-han Park, Soo-young Shin, "Implementation of Fall Detection Based on CNN-LSTM," The Journal of Korean Institute of Communications and Information Sciences, vol. 47, no. 2, pp. 340-347, 2. 2022. (https://doi.org/10.7840/kics.2022.47.2.340)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880