Share Email Print
cover

Proceedings Paper

Can state-of-the-art HVS-based objective image quality criteria be used for image reconstruction techniques based on ROI analysis?
Author(s): P. Dostal; L. Krasula; M. Klima
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Various image processing techniques in multimedia technology are optimized using visual attention feature of the human visual system. Spatial non-uniformity causes that different locations in an image are of different importance in terms of perception of the image. In other words, the perceived image quality depends mainly on the quality of important locations known as regions of interest. The performance of such techniques is measured by subjective evaluation or objective image quality criteria. Many state-of-the-art objective metrics are based on HVS properties; SSIM, MS-SSIM based on image structural information, VIF based on the information that human brain can ideally gain from the reference image or FSIM utilizing the low-level features to assign the different importance to each location in the image. But still none of these objective metrics utilize the analysis of regions of interest. We solve the question if these objective metrics can be used for effective evaluation of images reconstructed by processing techniques based on ROI analysis utilizing high-level features. In this paper authors show that the state-of-the-art objective metrics do not correlate well with subjective evaluation while the demosaicing based on ROI analysis is used for reconstruction. The ROI were computed from "ground truth" visual attention data. The algorithm combining two known demosaicing techniques on the basis of ROI location is proposed to reconstruct the ROI in fine quality while the rest of image is reconstructed with low quality. The color image reconstructed by this ROI approach was compared with selected demosaicing techniques by objective criteria and subjective testing. The qualitative comparison of the objective and subjective results indicates that the state-of-the-art objective metrics are still not suitable for evaluation image processing techniques based on ROI analysis and new criteria is demanded.

Paper Details

Date Published: 1 May 2012
PDF: 12 pages
Proc. SPIE 8436, Optics, Photonics, and Digital Technologies for Multimedia Applications II, 84361K (1 May 2012); doi: 10.1117/12.923612
Show Author Affiliations
P. Dostal, Czech Technical Univ. in Prague (Czech Republic)
L. Krasula, Czech Technical Univ. in Prague (Czech Republic)
M. Klima, Czech Technical Univ. in Prague (Czech Republic)


Published in SPIE Proceedings Vol. 8436:
Optics, Photonics, and Digital Technologies for Multimedia Applications II
Peter Schelkens; Touradj Ebrahimi; Gabriel Cristóbal; Frédéric Truchetet; Pasi Saarikko, Editor(s)

© SPIE. Terms of Use
Back to Top