Subjective image quality assessment optimization

Abstract

As the results of computer algorithms methods are often visual, image quality assessment is one of its central problems. To provide a convincing proof that a new method is better than the state-of-theart the image quality assessment should be employed. Therefore image based projects are often accompanied by user studies, in which a group of observers rank or rate results of several algorithms. Unfortunately the problem posed by subjective experiments is their time-consuming and expensive nature. This paper is intended to present how to make the subjective experiments less expensive and therefore more usable.