Blind deblurring with patch-wise second-order gradient prior

作者：

Highlights：

• Propose a new image prior and present an efficient blind deblurring model.

• Kernel similarity constraint is used to control the iteration times.

• Our approach can perform better than the state-of-the-art methods on various challenging scenes.

摘要

•Propose a new image prior and present an efficient blind deblurring model.•Kernel similarity constraint is used to control the iteration times.•Our approach can perform better than the state-of-the-art methods on various challenging scenes.

论文关键词：Motion deblurring,Second-order gradient,Kernel similarity,Blind deconvolution,Local image patches

论文评审过程：Received 17 November 2021, Revised 7 June 2022, Accepted 13 June 2022, Available online 17 June 2022, Version of Record 5 July 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116781