•   Home
  • Proceedings
  • International Conference on Information Technology and Scien...

International Conference on Information Technology and Scientific Management (ICITSM 2010 E-BOOK)

Tianjin,China,12.20-12.21,2010

ISBN: 978-1-935068-40-2 Scientific Research Publishing, USA

E-Book 1078pp Pub. Date: December 2010

Category: Computer Science & Communications

Price: $220

Title: Identification of Body Types Using KFDA in Non-Contacted Body Measurement System
Source: International Conference on Information Technology and Scientific Management (ICITSM 2010 E-BOOK) (pp 341-343)
Author(s): Yuxiu Wang, School of Art and Clothing, Tianjin Polytechnic University, Tianjin, China
Xiaojiu Li, School of Art and Clothing, Tianjin Polytechnic University, Tianjin, China
Hao Liu, School of textile, Tianjin Polytechnic University, Tianjin, China
Xiaozhi Li, School of Art and Clothing, Tianjin Polytechnic University, Tianjin, China
Abstract: Fast measurement and automatic classification of body types are critical parts in the automatic garment manufacture and made-to-measure (MTM). The calculating accuracy can be upgraded by using the clustering function to classify the body types and the clustering method is presented for classifying the body types without prior knowledge. This paper is focused on using the algorithm of kernel Fisher discriminant analysis (KFDA) to recognize automatically body type. The F-statistic is utilized for determining the optimal class number. KFDA, backpropagation neural network (BPNN) and radial basis function neural network (RBFNN) are compared in the Iris dataset and performed in the body dataset. Experimental results exhibit the KFDA is optimal method for identifying the body data. The result shows that calculating accuracy can be improved by using the discriminant analysis.
Proceedings by Subjects
Resources
Products
Contact us
book@scirp.org
WhatsApp +86 18163351462(WhatsApp)
Click here to send a message to me 1243940697
Book Publishing WeChat
Follow SCIRP
Free SCIRP Newsletters
Copyright © 2006-2024 Scientific Research Publishing Inc. All Rights Reserved.
Top