Skip to main navigation Skip to search Skip to main content

Pattern classification for dermoscopic images based on structure textons and Bag-of-Features model

  • Beihang University
  • Beijing Air Force General Hospital

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

Abstract

An effective method of pattern classification for dermoscopic images based on structure textons and Bag-of-Features (BoFs) model is proposed in this paper. Firstly, the pattern structures of images were enhanced. Secondly, images with obvious directivity were rotated to align their principal directions with horizontal axis, and Otsu method was used to obtain interesting regions. The intensity values of each pixel in the interesting region and its neighborhood composed patch vector. For each pattern, patch vectors of training images were clustered to generate K structure textons and a dictionary with 5K elements was obtained. Then BoFs model was applied to obtain texton histograms for training and testing images respectively. Finally, a nearest neighbor classifier with chi-square distance was adopted to classify. The experimental results shows that our enhancement method is beneficial to pattern classification and correct classification rate achieves 91.87%.

Original languageEnglish
Title of host publicationImage and Graphics - 8th International Conference, ICIG 2015, Proceedings, Part III
EditorsYu-Jin Zhang
PublisherSpringer Verlag
Pages34-45
Number of pages12
ISBN (Print)9783319219684
DOIs
StatePublished - 2015
Event8th International Conference on Image and Graphics, ICIG 2015 - Tianjin, China
Duration: 13 Aug 201516 Aug 2015

Publication series

NameLecture Notes in Computer Science
Volume9219
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference8th International Conference on Image and Graphics, ICIG 2015
Country/TerritoryChina
CityTianjin
Period13/08/1516/08/15

Keywords

  • Computer-aided diagnosis
  • Dermoscopic image
  • Pattern classification
  • Texton

Fingerprint

Dive into the research topics of 'Pattern classification for dermoscopic images based on structure textons and Bag-of-Features model'. Together they form a unique fingerprint.

Cite this