Spatially weighted order binary pattern for color texture classification

作者：

Highlights：

• We propose a novel texture descriptor to handle color images.

• A multi-channel color order pattern is explored to jointly encode inter-channel features.

• A discriminative color gradient channel is introduced to complement the RGB channels.

• Local color differences are decomposed to encode the spatial relationships of color orders.

• The proposed method achieves the state-of-the-art results for color texture classification.

摘要

•We propose a novel texture descriptor to handle color images.•A multi-channel color order pattern is explored to jointly encode inter-channel features.•A discriminative color gradient channel is introduced to complement the RGB channels.•Local color differences are decomposed to encode the spatial relationships of color orders.•The proposed method achieves the state-of-the-art results for color texture classification.

论文关键词：Color,Features,Local Binary Pattern (LBP),Texture classification

论文评审过程：Received 12 October 2019, Revised 6 December 2019, Accepted 27 December 2019, Available online 22 January 2020, Version of Record 27 January 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2019.113167