Development of New Quality Metric for Assessment of Decolorization

Abstract

Decolorization also well-known as color2gray image conversion is useful for monotone display and printing of images. Meaningful structure may vanish when a color image is transformed to a gray-scale image. Despite of its wide real world applications, adequate efforts has been made on image quality metric that can be correlated with perceived quality of images. Subjective quality evaluation is time consuming and is susceptible to personal bias thus objective quality perception model has been suggested in this research paper. This research proposes an algorithm consisting of three quality metrics, i.e. local complexity measure, contrast and edge density. The three metrics are integrated to yield an image quality assessment results. The experiments performed on benchmark images shows that the proposed algorithm evaluates images on the basis of content preservation and true contrast loss. The method is unique in its ability to measure image quality based on local, glocal and global level separately and mirrors multi-resolution analysis of visual data. It is first attempt among many to develop objective quality measurement model with these metrics that quantitatively evaluates gray scale image results.