Learning salient self-representation for image recognition via orthogonal transformation

作者：

Highlights：

• We propose salient self-representation (S2R) that learns the salient information.

• A S2R-based classifier was specially designed for pattern classification.

• We prove the rationale of using salient information for pattern classification.

• Extensive experiments were performed to show the effectiveness of S2R.

摘要

•We propose salient self-representation (S2R) that learns the salient information.•A S2R-based classifier was specially designed for pattern classification.•We prove the rationale of using salient information for pattern classification.•Extensive experiments were performed to show the effectiveness of S2R.

论文关键词：Pattern classification,Self-representation,Orthogonal transformation,Linear orthogonal transformation

论文评审过程：Received 27 January 2022, Revised 8 August 2022, Accepted 21 August 2022, Available online 29 August 2022, Version of Record 6 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118663