Driver Drowsiness Detection Based on Visual-Feature Using Multi-Modal Learning 


Vol. 43,  No. 7, pp. 1124-1132, Jul.  2018
10.7840/kics.2018.43.7.1124


PDF
  Abstract

There is a vision-based drowsiness detection method for judging drowsiness in a limited situation of driver. The driver"s eye and head movements are used to determine the drowsiness. In the case of drowsiness, the change in drowsiness is different. Since the drowsiness is different from other drowsiness, This may be difficult. In this paper, we propose a multimodal learning method using DBM(Deep Boltzmann Machine) in order not to be affected by visual feature data. Two of the commonly used visual-based features (eye, mouth) are used to observe changes in drowsiness. First, we study two feature data using multimodal DBM and then consider the temporal change using RNN (Recurrent Neural Network).

  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]

H. Choi, M. Back, J. Kang, K. Lee, "Driver Drowsiness Detection Based on Visual-Feature Using Multi-Modal Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 7, pp. 1124-1132, 2018. DOI: 10.7840/kics.2018.43.7.1124.

[ACM Style]

Hyung-tak Choi, Moon-ki Back, Jae-sik Kang, and Kyu-chul Lee. 2018. Driver Drowsiness Detection Based on Visual-Feature Using Multi-Modal Learning. The Journal of Korean Institute of Communications and Information Sciences, 43, 7, (2018), 1124-1132. DOI: 10.7840/kics.2018.43.7.1124.

[KICS Style]

Hyung-tak Choi, Moon-ki Back, Jae-sik Kang, Kyu-chul Lee, "Driver Drowsiness Detection Based on Visual-Feature Using Multi-Modal Learning," The Journal of Korean Institute of Communications and Information Sciences, vol. 43, no. 7, pp. 1124-1132, 7. 2018. (https://doi.org/10.7840/kics.2018.43.7.1124)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880