Classification of textile image using support vector machine with textural feature

Ratri Enggar Pawening, Rohman Dijaya, Thomas Brian, Nanik Suciati

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

25 Citations (Scopus)

Abstract

Fabrics have different materials, colors, and texture. The development of fashion is influenced by climate and fashion trends. In the particular climate of the summer for example, a floral pattern clothes are interesting and in other climates, the other fashion patterns, such as lines, polka dot, and strips might be more interesting. This study focuses on textile image classification based on its texture. Each texture in textile image has a particular characteristic that can distinguish with other motifs. The feature extraction method used in this study are Gray Level Co-occurrence Matrix (GLCM), Linear Binary Pattern (LBP), and a Moment Invariant (MI). Furthermore, all texture feature are then reduced using Principal Component Analysis (PCA). The experiment shows that the best result can be achieved using combination of GLCM and LBP features with accuracy 74.15% using linear kernel SVM.

Original languageEnglish
Title of host publicationProceedings of 2015 International Conference on Information and Communication Technology and Systems, ICTS 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages119-122
Number of pages4
ISBN (Electronic)9781509000951
DOIs
Publication statusPublished - 12 Jan 2016
EventInternational Conference on Information and Communication Technology and Systems, ICTS 2015 - Surabaya, Indonesia
Duration: 16 Sept 2015 → …

Publication series

NameProceedings of 2015 International Conference on Information and Communication Technology and Systems, ICTS 2015

Conference

ConferenceInternational Conference on Information and Communication Technology and Systems, ICTS 2015
Country/TerritoryIndonesia
CitySurabaya
Period16/09/15 → …

Keywords

  • GLCM
  • LBP
  • SVM
  • classification
  • feature texture
  • moment invariant
  • textile

Fingerprint

Dive into the research topics of 'Classification of textile image using support vector machine with textural feature'. Together they form a unique fingerprint.

Cite this