达哥是超人的博客

Note：Anchored NeighborhoodRegression for Fast Example-Based Super-Resolution 1.        Abstracta)        Propose fast super-resolutionmethods while making no compromise on quality

Xue W, Zhang L, Mou X. Learning without human scores for blindimage quality assessment[C]//Proceedings of the IEEE Conference on ComputerVision and Pattern Recognition. 2013: 995-1002.1.        Abst

1.  主观评价方法1)        准备样本集2)        由观测者对样本集图像进行评价3)        对主观评价结果进行加工2.   客观评价方法1)        全参考a)        基于误差统计量的算法                        i.             来源于通信理论：信号保真度（峰值信噪比、目标信息密度）