PROJECT TITLE :

A Detail-Based Method for Linear Full Reference Image Quality Prediction - 2018

ABSTRACT:

During this Project, a completely unique full Reference technique is proposed for image quality assessment, using the mixture of 2 separate metrics to measure the perceptually distinct impact of detail losses and of spurious details. To this purpose, the gradient of the impaired image is domestically decomposed as a predicted version of the original gradient, plus a gradient residual. It's assumed that the detail attenuation identifies the detail loss, whereas the gradient residuals describe the spurious details. It seems that the perceptual impact of detail losses is roughly linear with the loss of the positional Fisher data, while the perceptual impact of the spurious details is roughly proportional to a logarithmic live of the signal to residual ratio. The affine combination of those 2 metrics forms a brand new index strongly correlated with the empirical differential mean opinion score (DMOS) for a significant category of image impairments, as verified for 3 freelance standard databases. The method allowed alignment and merging of DMOS information coming from these different databases to a common DMOS scale by affine transformations. Unexpectedly, the DMOS scale setting is possible by the analysis of one image suffering from additive noise.