Quality Assessment of Resampled Digital Images by Statistical Metrics
Author(s) -
W. Korneta
Publication year - 2017
Publication title -
edukacja – technika – informatyka
Language(s) - English
Resource type - Journals
eISSN - 2450-9221
pISSN - 2080-9069
DOI - 10.15584/eti.2017.3.40
Subject(s) - grayscale , luminance , contrast (vision) , image quality , artificial intelligence , index (typography) , computer science , pattern recognition (psychology) , statistics , computer vision , mathematics , image (mathematics) , world wide web
The visual degradation of resampled (downsampled and then upsampled to the original size) greyscale digital images is quantified by local luminance, contrast and structure statistical comparison indexes. Spatial distributions of these indexes are shown. The global quality of resampled image is quantifiedby the similarity index defined by median values of local indexes. Parameters of this index consistent with the median opinion score are determined. The dependence of global indexes on the size of downsampled image is presented.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom