A1 Journal article (refereed), original research

Full Reference Printed Image Quality: Measurement Framework and Statistical Evaluation


Open Access publication

Publication Details
Authors: Eerola Tuomas, Kämäräinen Joni-Kristian, Lensu Lasse, Leisti Tuomas, Halonen Raisa, Kälviäinen Heikki, Nyman Göte, Oittinen Pirkko
Publication year: 2010
Language: English
Related Journal or Series Information: Journal of Imaging Science and Technology
Volume number: 54
Issue number: 1
ISSN: 1062-3701
JUFO-Level of this publication:
Open Access: Open Access publication

Abstract
Full reference image quality algorithms are standard tools in digital image processing but have not been utilized for printed images due to a "correspondence gap" between the digital domain (a reference) and physical domain (printed sample). In this work, the authors propose a framework for applying full reference image quality algorithms to printed images. The framework consists of accurate scanning of printed samples and automatic registration and descreening procedures which bring the scans in correspondence with their digital originals. The authors complete the framework by incorporating state-of-the-art full reference algorithms to it. Using data from comprehensive psychometrical experiments of subjective quality experience, the authors benchmark the state-of-theart methods and point out similar results in the digital domain: the best digital full reference measures, such as the recently introduced visual information fidelity algorithm, perform best also for printed media.

Last updated on 2017-22-03 at 14:44