Full Reference Image Quality Metrics and their Performance

Kipli, K. and Krishnan, Shekhar and Zamhari, N. and Muhammad, M.S. and Masra, S.M.W. and Kho, Lee Chin and Lias, K. (2011) Full Reference Image Quality Metrics and their Performance. IEEE 7th International Colloquium on Signal Processing and its Applications (CSPA), 2011. ISSN Print ISBN: 978-1-61284-414-5

[img]
Preview
PDF
Full Reference Image Quality Metrics and their PERFORMANCE (abstract).pdf

Download (414kB) | Preview
Official URL: http://ieeexplore.ieee.org/document/5759838/

Abstract

This paper mainly aims to study the performance of objective assessment methods of image quality. It take into consideration the correlations between each objective assessment and the subjective assessment in order to determine objective test performance. Three objective assessment methods used in this study are the Structural Similarity (SSIM) index, the Peak Signal-to-Noise Ratio (PSNR) and the Mean Squared Error (MSE) calculating algorithm. The resulting data indicate what type of objective assessment was most suitable for which type of impairment imposed upon an image. This is clarified using the Pearson Correlation Coefficient as described in the paper. As an overall, SSIM index had the best correlation characteristics to the subjective assessment, followed by the MSE calculating algorithm. From this study, a better understanding of the requirements for developing an efficient image quality assessment method was gained.

Item Type: Article
Uncontrolled Keywords: Image Quality Metrics, PSNR, SSIM, MSE, unimas, university, universiti, Borneo, Malaysia, Sarawak, Kuching, Samarahan, ipta, education, research, Universiti Malaysia Sarawak
Subjects: T Technology > T Technology (General)
Divisions: Academic Faculties, Institutes and Centres > Faculty of Engineering
Faculties, Institutes, Centres > Faculty of Engineering
Depositing User: Karen Kornalius
Date Deposited: 27 Sep 2017 07:14
Last Modified: 27 Sep 2017 07:14
URI: http://ir.unimas.my/id/eprint/17808

Actions (For repository members only: login required)

View Item View Item