Geometrical Feature Extraction of Hidden Objects with Millimeter Wave Imaging by Shape Descriptors and Principal Components 


Vol. 42,  No. 10, pp. 1878-1885, Oct.  2017
10.7840/kics.2017.42.10.1878


PDF
  Abstract

Millimeter wave (MMW) imaging has found rapid adoption in security applications such as concealed object detection because of is penetrating property to clothing. This study addresses the concealed object segmentation and geometric feature extraction with passive MMW imaging. The multi-level segmentation extracts the body area at the first level and the object area from the body area at the second level. Shape descriptors such as perimeter, size, major and minor axes, and the eigenvalues of the principal components are calculated from the segmented object area. Scale, rotation, and translation invariant features are composed of compactness, rectangularity, ellipse eccentricity, and principal component eccentricity. In the experiments, three metallic objects (gun, knife, hand-axe) and one non-metallic objects (plastic lotion bottle containing liquid) concealed under clothing are captured by the passive MMW imaging system. The geometric features extracted from the segmented areas and the binary models of the real objects are compared showing more than 91% accuracy for metallic objects and 87% for a non metallic object.

  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]

S. Yeom and J. Son, "Geometrical Feature Extraction of Hidden Objects with Millimeter Wave Imaging by Shape Descriptors and Principal Components," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 10, pp. 1878-1885, 2017. DOI: 10.7840/kics.2017.42.10.1878.

[ACM Style]

Seokwon Yeom and Jung-young Son. 2017. Geometrical Feature Extraction of Hidden Objects with Millimeter Wave Imaging by Shape Descriptors and Principal Components. The Journal of Korean Institute of Communications and Information Sciences, 42, 10, (2017), 1878-1885. DOI: 10.7840/kics.2017.42.10.1878.

[KICS Style]

Seokwon Yeom and Jung-young Son, "Geometrical Feature Extraction of Hidden Objects with Millimeter Wave Imaging by Shape Descriptors and Principal Components," The Journal of Korean Institute of Communications and Information Sciences, vol. 42, no. 10, pp. 1878-1885, 10. 2017. (https://doi.org/10.7840/kics.2017.42.10.1878)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880