A new objective, full-reference metrics of image quality is proposed in this
paper. It should match perceptual (subjective) image quality assessment in a
better way. The proposed method consists of two quality measures which
separately indicate image quality on edges and in texture areas which are
calculated in a three-step algorithm. The ?soft mask? is initially found for
separation in edge and texture areas. Then, two MSEs (mean square error) with
corresponding two PSNRs (peak signal-to-noise ratio) for edge and texture are
calculated using soft mask as the weighting factor. Finally, the obtained two
PSNRs are re-calculated into the two quality indices for edges and texture.
Additionally, the separation factor, defined as percentage of edge areas in
image, is considered, describing the influence of the image content on
perceptual assessment. The proposed 2D metrics is especially suited for
evaluations of different interpolation and compression algorithms.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.