A study of the most important image quality measures

Abstract

At the present time there are several statistical computer measures in existence that can be used to quantify the gain or loss of image quality. The challenge consists in guaranteeing that these measures may be reliable for a given goal. This paper presents a study and comparison between several commonly used quality measures (signal/noise relation, average quadratic error, histogram cross-correlation, and pixel correlation) applied to digital images that underwent computer processing to affect their quality.