Smoke Detection Algorithm Using Deep Learning 


Vol. 42,  No. 7, pp. 1370-1379, Jul.  2017
10.7840/kics.2017.42.7.1370


PDF
  Abstract

Fire is a fatal accident that causes personal injury and property damage. Therefore, rapid fire detection and warning are the best ways to reduce fire damage. Detecting smoke from an early fire is the most important clue in a fire alarm system. In this paper, we propose an automatic smoke detection algorithm based on camera surveillance system and image processing technology. In this algorithm, smoke is detected and traced to a moving object, and then a cascade classification model using CNN (Convolutional Neural Network) is used to distinguish smoke from non-smoke objects. The results of this study show good results in detecting smoke as well as reducing false alarms when compared with existing studies.

  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]

N. M. Dung and S. Ro, "Smoke Detection Algorithm Using Deep Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 7, pp. 1370-1379, 2017. DOI: 10.7840/kics.2017.42.7.1370.

[ACM Style]

Nguyen Manh Dung and Soonghwan Ro. 2017. Smoke Detection Algorithm Using Deep Learning. The Journal of Korean Institute of Communications and Information Sciences, 42, 7, (2017), 1370-1379. DOI: 10.7840/kics.2017.42.7.1370.

[KICS Style]

Nguyen Manh Dung and Soonghwan Ro, "Smoke Detection Algorithm Using Deep Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 7, pp. 1370-1379, 7. 2017. (https://doi.org/10.7840/kics.2017.42.7.1370)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880