Presentation 2014-03-06
Multifractal computation based feature descriptor for texture classification
Chamidu Atupelage, Hiroshi Nagahashi,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper describes a textural feature extraction based on multifractal computation. Multifractal is used to measure the complexity of natural texture. Using multifractal computation, we derived a high-dimensional feature space to describe the pixels' local and spatial distribution characteristics. Our. feature descriptor utilizes four multifractal measures that are invariant for the rotational changes of the texture. We investigated the performance of multifractal feature descriptor by comparing the classification accuracies of three popular feature descriptors. The results indicate that multifractal features contain higher discrimination capacity.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Texture classification / Textural feature descriptors / Fractal / Multifractal
Paper # IMQ2013-44,IE2013-153,MVE2013-82
Date of Issue

Conference Information
Committee MVE
Conference Date 2014/2/27(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Media Experience and Virtual Environment (MVE)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Multifractal computation based feature descriptor for texture classification
Sub Title (in English)
Keyword(1) Texture classification
Keyword(2) Textural feature descriptors
Keyword(3) Fractal
Keyword(4) Multifractal
1st Author's Name Chamidu Atupelage
1st Author's Affiliation Imaging Science and Engineering Laboratory, Tokyo Institute of Technology()
2nd Author's Name Hiroshi Nagahashi
2nd Author's Affiliation Imaging Science and Engineering Laboratory, Tokyo Institute of Technology
Date 2014-03-06
Paper # IMQ2013-44,IE2013-153,MVE2013-82
Volume (vol) vol.113
Number (no) 470
Page pp.pp.-
#Pages 5
Date of Issue