Fabric defect detection based on completed local quartet patterns and majority decision algorithm

作者：

Highlights：

• Fabric textile defect detection approach based on completed local quartet patterns and majority decision algorithm.

• A new local texture descriptor called completed local quartet patterns is presented to extract discriminative texture features.

• An innovative majority voting algorithm is proposed to improve final detection accuracy.

• The proposed approach has high accuracy at or above 97.2 percent in patterned fabrics with different pattern types.

摘要

•Fabric textile defect detection approach based on completed local quartet patterns and majority decision algorithm.•A new local texture descriptor called completed local quartet patterns is presented to extract discriminative texture features.•An innovative majority voting algorithm is proposed to improve final detection accuracy.•The proposed approach has high accuracy at or above 97.2 percent in patterned fabrics with different pattern types.

论文关键词：Visual inspection system,Surface defect detection,Texture analysis,Fabric textile defects,Completed local quartet patterns,Majority decision algorithm

论文评审过程：Received 28 July 2020, Revised 16 October 2021, Accepted 25 February 2022, Available online 9 March 2022, Version of Record 15 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116827