Image quality metric based on multidimensional contrast perception models

摘要

The procedure to compute the subjective difference between two input images should be equivalent to a straightforward difference between their perceived versions, hence reliable subjective difference metrics mush be founded on a proper perception model. For image distortion evaluation purpose, perception can be considered as a set of signal transform that successively map the original Image in the spatial domain into a feature and a response space. The properties of the spatial pattern analyzers involved in these transforms Determine the geometry of these different signal representation domains.