Microscopic Local Binary Pattern for Texture Classification

Jiangping HE  Wei SONG  Hongwei JI  Xin YANG  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E95-A   No.9   pp.1587-1595
Publication Date: 2012/09/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E95.A.1587
Print ISSN: 0916-8508
Type of Manuscript: PAPER
Category: Image
texture analysis,  local binary pattern,  microscopic level,  structure information,  

Full Text: PDF>>
Buy this Article

This paper presents a Microscopic Local Binary Pattern (MLBP) for texture classification. The conventional LBP methods which rely on the uniform patterns discard some texture information by merging the nonuniform patterns. MLBP preserves the information by classifying the nonuniform patterns using the structure similarity at microscopic level. First, the nonuniform patterns are classified into three groups using the macroscopic information. Second, the three groups are individually divided into several subgroups based on the microscopic structure information. The experiments show that MLBP achieves a better result compared with the other LBP related methods.