Scale invariant texture analysis using multi-scale local autocorrelation features

Yousun Kang, Ken'ichi Morooka, Hiroshi Nagahashi

Research output: Contribution to journalConference articlepeer-review

10 Citations (Scopus)


We have developed a new framework for scale invariant texture analysis using multi-scale local autocorrelation features. The multi-scale features are made of concatenated feature vectors of different scales, which are calculated from higher-order local autocorrelation functions. To classify different types of textures among the given test images, a linear discriminant classifier (LDA) is employed in the multi-scale feature space. The scale rate of test patterns in their reduced subspace can also be estimated by principal component analysis (PCA). This subspace represents the scale variation of each scale step by principal components of a training texture image. Experimental results show that the proposed method is effective in not only scale invariant texture classification including estimation of scale rate, but also scale invariant segmentation of 2D image for scene analysis.

Original languageEnglish
Pages (from-to)363-373
Number of pages11
JournalLecture Notes in Computer Science
Publication statusPublished - 2005
Externally publishedYes
Event5th International Conference on Scale Space and PDE Methods in Computer Vision, Scale-Space 2005 - Hofgeismar, Germany
Duration: Apr 7 2005Apr 9 2005

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Scale invariant texture analysis using multi-scale local autocorrelation features'. Together they form a unique fingerprint.

Cite this