Method of Fatigue Driving State Automatic Detection based on Dynamic Image

  • Cheng Ya-ling, Tan Ai-ping
Keywords: Dynamic Image; Eye State; Kurtosis; Skewness; Fatigue Driving

Abstract

Most of the traditional fatigue driving detection methods process each frame separately, to make full use of the time information of video sequence, a method of fatigue driving state detection based on dynamic image is proposed. Firstly, generate a dynamic image corresponding to face video of the driver. Then, the regions of eyes are extracted by image segmentation method. Finally, the kurtosis and skewness values of the gray distribution of the eyes’ regions are calculated, and the presence or absence of fatigue in the video sequence is determined according to the normality of the gray distribution of the eyes’ regions. In the experiment, a section of face video data containing multiple sets of normal and fatigue states was analyzed. The results show that our method can accurately distinguish the blink behaviors under normal conditions from those under fatigue, which verified the effectiveness of the proposed method.

Published
2021-07-29
How to Cite
Tan Ai-ping, C. Y.- ling,. (2021). Method of Fatigue Driving State Automatic Detection based on Dynamic Image. Design Engineering, 5444- 5454. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/3004
Section
Articles