Grayscale image quality measure in spatial domain

Measurement of image quality is of fundamental importance to many image and video processing applications. Two approaches for image quality measurement are subjective and objective. Subjective measurement is concerned with result of human expert giving their opinion of image quality where as objective measurements are done with the aid of mathematical algorithms. Existing objective quality metrics like MSSIM which is based on structure similarity, fail on particular image impairments like in case of highly blurred and compressed images. Perceived image distortion not only depends upon structure but also on local features such as edges, flats, and texture. In this paper a new index for image quality assessment has been proposed. This proposed index is applied on images distorted with a combination of three factors: structure distortion, edge distortion, and luminance distortion. Proposed index is mathematically defined and in it HVS model is explicitly employed, our experiments on various image distortion types indicate that this index performs significantly better than traditional error summation methods and other existing structure similarity measures.