Non-uniform low-light image enhancement via non-local similarity decomposition model

作者：

Highlights：

• A non-local similarity decomposition model is proposed for enhancing nonuniform low-light images.

• The edge information is used to constrain the gradient of illumination.

• The SM is defined, which represents intrinsic characteristics, including color, contrast and texture.

• Non-local similarity is utilized to constrain the reflectance component.

摘要

•A non-local similarity decomposition model is proposed for enhancing nonuniform low-light images.•The edge information is used to constrain the gradient of illumination.•The SM is defined, which represents intrinsic characteristics, including color, contrast and texture.•Non-local similarity is utilized to constrain the reflectance component.

论文关键词：Non-uniform low-light image enhancement,Similarity measure,Non-local similarity decomposition,Edge information,Reflectance and illumination

论文评审过程：Received 29 January 2020, Revised 2 December 2020, Accepted 4 January 2021, Available online 13 January 2021, Version of Record 30 January 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116141