Structural similarity index (SSIM) revisited: A data-driven approach

作者：

Highlights：

• Optimization of the parameters of quality assessment measures.

• Definition of a new set of reference parameters for SSIM.

• Results outperforms the state-of-the-art techniques.

• Parameters can be successfully transferred to different datasets.

• Single and multi-scale structural similarity index are considered.

摘要

•Optimization of the parameters of quality assessment measures.•Definition of a new set of reference parameters for SSIM.•Results outperforms the state-of-the-art techniques.•Parameters can be successfully transferred to different datasets.•Single and multi-scale structural similarity index are considered.

论文关键词：Image quality assessment measures,Structural similarity,Evolutionary computation,Scale selection,Image processing

论文评审过程：Received 27 August 2020, Revised 2 August 2021, Accepted 12 October 2021, Available online 27 October 2021, Version of Record 3 November 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116087