Down syndrome recognition using local binary patterns and statistical evaluation of the system

Burcin K., Vasif N. V.

EXPERT SYSTEMS WITH APPLICATIONS, vol.38, no.7, pp.8690-8695, 2011 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 38 Issue: 7
  • Publication Date: 2011
  • Doi Number: 10.1016/j.eswa.2011.01.076
  • Journal Name: EXPERT SYSTEMS WITH APPLICATIONS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.8690-8695
  • Keywords: Down syndrome recognition, Local binary pattern, Feature extraction, Classification, CLASSIFICATION
  • Karadeniz Technical University Affiliated: Yes

Abstract

Down syndrome has a private facial view, thus it can be recognized by using facial features. But this is a very challenging problem when the similarity between the faces of people with Down syndrome and not Down syndrome people are considered. Therefore, we used the local binary pattern (LBP) approach for feature extraction which is a very effective feature descriptor. For classification Euclidean distance and Changed Manhattan distance methods are used. In this way, we improved an efficient system to recognize Down syndrome. (C) 2011 Elsevier Ltd. All rights reserved.