Globally rotation invariant multi-scale co-occurrence local binary pattern

Xianbiao Qi, Linlin Shen, Guoying Zhao, Qingquan Li, Matti Pietikäinen

Research output: Journal PublicationArticlepeer-review

33 Citations (Scopus)

Abstract

This paper proposes a globally rotation invariant multi-scale co-occurrence local binary pattern (MCLBP) feature for texture-relevant tasks. In MCLBP, we arrange all co-occurrence patterns into groups according to properties of the co-patterns, and design three encoding functions (Sum, Moment, and Fourier Pooling) to extract features from each group. The MCLBP can effectively capture the correlation information between different scales and is also globally rotation invariant (GRI). The MCLBP is substantially different from most existing LBP variants including the LBP, the CLBP, and the MSJ-LBP that achieves rotation invariance by locally rotation invariant (LRI) encoding. We fully evaluate the properties of the MCLBP and compare it with some powerful features on five challenging databases. Extensive experiments demonstrate the effectiveness of the MCLBP compared to the state-of-the-art LBP variants including the CLBP and the LBPHF. Meanwhile, the dimension and computational cost of the MCLBP is also lower than that of the CLBP-S/M/C and LBPHF-S-M.

Original languageEnglish
Pages (from-to)16-26
Number of pages11
JournalImage and Vision Computing
Volume43
DOIs
Publication statusPublished - 9 Nov 2015
Externally publishedYes

Keywords

  • Globally rotation invariant
  • Locally rotation invariant
  • Multi-scale co-occurrence LBP
  • Texture classification

ASJC Scopus subject areas

  • Signal Processing
  • Computer Vision and Pattern Recognition

Fingerprint

Dive into the research topics of 'Globally rotation invariant multi-scale co-occurrence local binary pattern'. Together they form a unique fingerprint.

Cite this