A new no-reference image quality measure to determine the quality of a given image using object separability

Abstract

The goal of researchers in the field of Image Quality Assessment is to quantify quality of an image using a mathematical measure and to design algorithms for computing the measure. The traditional method of doing this involves taking a reference image and a test image of same scene and find differences between the two images. As human eye can differentiate between a good quality image and a distorted one without the use of reference image, in this paper we propose a no-reference image quality measure which will differentiate between a good image and distorted image by calculating certain properties of images based on object separability in the image.