Improving Personal Protective Equipment Detection in Smart Construction 


Vol. 45,  No. 12, pp. 2202-2209, Dec.  2020
10.7840/kics.2020.45.12.2202


PDF
  Abstract

In order to reduce accidents and fatalities in construction sites, it is important to ensure that the workers are wearing appropriate personal protective equipments (PPEs). Manual inspection and monitoring of PPE usage is time consuming and requires significant human resources. The introduction of smart construction enabled object detection techniques to be used for automatic detection of PPE usage. Among various object detection algorithms, YOLO enables PPE detection in real-time and is widely used. If the monitoring camera is installed afar in order to secure a wide view of the construction site, the size of PPE objects in the output image becomes very small. YOLO has weaknesses in detecting small objects, so in this case, it will not be able to perform effective PPE detection. In this paper, a motion mosaic image pre-processing method is proposed to improve small PPE detection by YOLO. The method usefs motion detection in order to extract areas of probable PPE locations. A mosaic image is created from the extracted areas and fed as input to YOLO. This method preserves the detection area resolution as best as possible, which increases small sized PPE detection rate by YOLO.

  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. Park, S. Lee, K. Lee, "Improving Personal Protective Equipment Detection in Smart Construction," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 12, pp. 2202-2209, 2020. DOI: 10.7840/kics.2020.45.12.2202.

[ACM Style]

Yong-Suk Park, Song-Yeon Lee, and Kyung-Taek Lee. 2020. Improving Personal Protective Equipment Detection in Smart Construction. The Journal of Korean Institute of Communications and Information Sciences, 45, 12, (2020), 2202-2209. DOI: 10.7840/kics.2020.45.12.2202.

[KICS Style]

Yong-Suk Park, Song-Yeon Lee, Kyung-Taek Lee, "Improving Personal Protective Equipment Detection in Smart Construction," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 12, pp. 2202-2209, 12. 2020. (https://doi.org/10.7840/kics.2020.45.12.2202)