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

作者：

Highlights：

• This paper proposes a globally rotation invariant multi-scale co-occurrence of LBPs (MCLBP).

• The proposed MCLBP can effectively capture the correlation between the LBPs in different scales.

• Three globally rotation invariant encoding methods are introduced for MCLBP.

• The proposed MCLBP performs very well on texture, material, and medical cell classification.

摘要

•This paper proposes a globally rotation invariant multi-scale co-occurrence of LBPs (MCLBP).•The proposed MCLBP can effectively capture the correlation between the LBPs in different scales.•Three globally rotation invariant encoding methods are introduced for MCLBP.•The proposed MCLBP performs very well on texture, material, and medical cell classification.

论文关键词：Multi-scale co-occurrence LBP,Globally rotation invariant,Locally rotation invariant,Texture classification

论文评审过程：Received 23 July 2014, Revised 29 May 2015, Accepted 28 July 2015, Available online 28 August 2015, Version of Record 8 September 2015.

论文官网地址：https://doi.org/10.1016/j.imavis.2015.07.005