本文为本人阅读论文《2018 PIRM Challenge on Perceptual Image Super-resolution》时写下的阅读笔记，看这篇paper主要是为了更深入的了解NIQE
本文提及，SR evaluation methodology jointly quanties accuracy and perceptual quality,（SR的评价体系应该联合量化准确度和感知质量）the more SR methods advanced, the more it has become evident that reconstruction accuracy and perceptual quality are typically in disagreement with each other.重建精度和感知质量通常彼此不一致。更少的重建的误差，会导致更差的视觉效果。
2018 PIRM challenge以知觉质量感知方式进行评价。define perceptual quality as the visual quality of the reconstructed image regardless of its similarity to any ground-truth image.
关于NIQE
https://blog.csdn.net/mazhitong1020/article/details/80415758
Natural image quality evaluator, NIQE——无需利用人眼评分的失真图像进行训练在计算其局部MSCN 归一化图像后, 根据局部活性选择部分图像块作为训练数据, 以广义高斯模型拟合得到模型参数作为特征, 采用多变量高斯模型描述这些特征, 评价过程中利用待评价图像特征模型参数与预先建立的模型参数之间的距离来确定图像质量