BGT: A blind image quality evaluator via gradient and texture statistical features

作者：

Highlights：

• We adopt conditional probability distribution to model the relationship between image gradients and textures.

• We design two sets of quality-aware features to evaluate image quality.

• We propose a high accuracy and excellent generalization properties BIQA model.

摘要

•We adopt conditional probability distribution to model the relationship between image gradients and textures.•We design two sets of quality-aware features to evaluate image quality.•We propose a high accuracy and excellent generalization properties BIQA model.

论文关键词：Blind image quality assessment (BIQA),Human visual system (HVS),Natural scene statistics (NSS),Gradient,Texture,Joint statistics

论文评审过程：Received 17 June 2020, Revised 29 April 2021, Accepted 3 May 2021, Available online 13 May 2021, Version of Record 17 May 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116315