• DocumentCode
    2998590
  • Title

    Algorithm of female´s non-contact somatometry system based on 2-dimensional pictures

  • Author

    Tan, Fei ; Xu, Xiao-Hui ; Liu, Guo-lian

  • Author_Institution
    Nat. Eng. Lab. for Modern Silk, Soochow Univ., Suzhou, China
  • fYear
    2009
  • fDate
    26-29 Nov. 2009
  • Firstpage
    1390
  • Lastpage
    1395
  • Abstract
    540 female were measured by using photo and manual measuring methods. Among these people there are 317 young and 223 elder. By using the Matlab program, female´s frontal and sided pictures were extracted and the outlines were got, from which the relevant height, thickness and width values could be drawn. Then construct the fitted equation of circumference by linear regression equation in two unknowns, through that the purpose of female´s non-contract technique of measurement based on 2-dimensional pictures can be realized. In order to improve the accuracy of measurement, the ratios of characteristic points´ height to body height in this system were divided while the ratio of thick to width was also divided, then for each grade, the algorithmic system were respectively established. Confirmatory analysis results show that the data got from the algorithmic system are credible.
  • Keywords
    clothing industry; mathematics computing; measurement; regression analysis; 2D pictures; Matlab program; female non-contact somatometry system; girth fitting; linear regression equation; manual measuring methods; Arm; Breast; Educational institutions; Equations; Hip; Humans; Laboratories; Neck; Thickness measurement; Velocity measurement; Girth fitting; Matlab; algorithmic; non-contact somatometry system; outline extraction;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer-Aided Industrial Design & Conceptual Design, 2009. CAID & CD 2009. IEEE 10th International Conference on
  • Conference_Location
    Wenzhou
  • Print_ISBN
    978-1-4244-5266-8
  • Electronic_ISBN
    978-1-4244-5268-2
  • Type

    conf

  • DOI
    10.1109/CAIDCD.2009.5375155
  • Filename
    5375155