Tensor-based plenoptic image denoising by integrating super-resolution

作者：

Highlights：

• Tensor-based plenoptic image denoising model is formulated by preserving structures.

• Similarity and diversity of the data in different dimensions are utilized.

• The ITS measure for MSI denoising is generalized for the plenoptic image denoising.

• Projection-based SR is integrated into denoising to eliminate the sub-pixel misalignment.

摘要

•Tensor-based plenoptic image denoising model is formulated by preserving structures.•Similarity and diversity of the data in different dimensions are utilized.•The ITS measure for MSI denoising is generalized for the plenoptic image denoising.•Projection-based SR is integrated into denoising to eliminate the sub-pixel misalignment.

论文关键词：Plenoptic image,Denoising,Super-resolution,Nonlocal similarity,Back-projection

论文评审过程：Received 31 August 2020, Revised 30 December 2021, Accepted 13 June 2022, Available online 22 June 2022, Version of Record 11 July 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116789