Image Quality Assessment: From Difference to Relative Entropy

Wachirapong Jirakitpuwapat, Poom Kumam, Tanapat Deesuwan, Anantachai Padcharoen, Duangkamon Kitkuan, Auwal Bala Abubakar

Abstract

Assessing image quality via objective methods basically attempts to quantify or measure the differences in visibility between a reference image and a distorted image. In this paper, a more general objective image quality index is given, which is simple to implement and has vast application in image processing. The relative entropy index is based on the assumption that the image is an information entropy. Experimental comparisons demonstrate the effectiveness of the proposed method.