A generalized quality assessment method for natural and screen content images

Abstract

A generalized objective quality assessment method is proposed for natural images andscreen content images. Since natural images and screen content images have different sta-tistical properties, the modelling of a generalized quality assessment method that worksfor both types of images is complicated because some properties of natural imag es andscreen content images are conflicting to one another. The proposed method assesses theperceptual quality of an image based on edge magnitude and direction. In this method, animage is first separated into regions with high and low gradients. Gradient is used due tothe small perceptual span of the human visual system for textual content. For high gradientregions, small kernel size of Prewitt operators is used to obtain the gradient magnitude anddirection. Correspondingly, bigger kernel size of Prewitt operators is utilized for low gra-dient regions. Visual quality indices are computed from both regions and pooled to obtainthe final quality index. From the performance comparison, it is shown that the proposedmethod could assess the perceived quality of natural images and screen content imageswith high accuracy.