A two-stage deep generative adversarial quality enhancement network for real-world 3D CT images

作者：

Highlights：

• We propose a two-stage deep network for the quality enhancement of 3D CT images.

• The quality enhancement is considered as unpaired image-to-image translation.

• The enhancement process benefits visual quality as well as property analysis.

• Higher quality images can be produced without upgrading imaging systems.

摘要

•We propose a two-stage deep network for the quality enhancement of 3D CT images.•The quality enhancement is considered as unpaired image-to-image translation.•The enhancement process benefits visual quality as well as property analysis.•Higher quality images can be produced without upgrading imaging systems.

论文关键词：Real-world 3D CT images,Quality enhancement,Generative adversarial networks,Morphological characteristics,Statistical properties,Rocks

论文评审过程：Received 30 April 2021, Revised 28 August 2021, Accepted 19 December 2021, Available online 6 January 2022, Version of Record 13 January 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116440