An Algorithm for Quality Assessment of Images with Contrast Changes and Mean Shifts

Abstract

The article presents the algorithm of an objective quality assessment of the images distorted by contrast change and mean shift. The stages of the development are considered. Full-reference image quality assessment algorithm and non- reference one are offered. The parameters of the assessment are obtained from the histograms of the original image and the distorted one. Also linear regression model is used to calculate the metric. Regression analysis methods are applied to determine effective subset of parameters.

Keywords

• computer vision
• image processing
• image quality