Target recognition in SAR images via sparse representation in the frequency domain

作者：

Highlights：

• A new sparse signal representation in the frequency domain is presented for target recognition in SAR image.

• A frequency prescreening skill is developed to generate representation in the transformed domain.

• The recognition performance of proposed method is verified in several unconstrained scenarios, including the change of pose, configuration, and depression angle, spatial translation, articulation and occlusion.

摘要

•A new sparse signal representation in the frequency domain is presented for target recognition in SAR image.•A frequency prescreening skill is developed to generate representation in the transformed domain.•The recognition performance of proposed method is verified in several unconstrained scenarios, including the change of pose, configuration, and depression angle, spatial translation, articulation and occlusion.

论文关键词：Sparse representation,Transformed domain,Target recognition

论文评审过程：Received 31 January 2019, Revised 4 July 2019, Accepted 15 July 2019, Available online 15 July 2019, Version of Record 19 July 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.106972