1 |
Evaluation Of Visual Quality MetricsOlgun, Ferhat Ramazan 01 October 2011 (has links) (PDF)
The aim of this study is to work on the visual quality metrics that are widely
accepted in literature, to evaluate them on different distortion types and to give a
comparison of overall performances in terms of prediction accuracy, monotonicity,
consistency and complexity. The algorithms behind the quality metrics in literature
and parameters used for quality metric performance evaluations are studied. This
thesis also includes the explanation of Human Visual System, classification of visual
quality metrics and subjective quality assessment methods. Experimental results that
show the correlation between objective scores and human perception are taken to
compare the eight widely accepted visual quality metrics.
|
Page generated in 0.12 seconds