For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Homogeneity Based Image Objective Quality Metric
Kebin AN Jun SUN Weina DU
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2006/10/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Image Processing and Video Processing
homogeneous regions, second order operators, image quality evaluation, peak signal to noise ratio (PSNR),
Full Text: PDF>>
A new fast and reliable image objective quality evaluation technique is presented in this paper. The proposed method takes image structure into account and uses a low complexity homogeneity measure to evaluate the intensity uniformity of a local region based on high-pass operators. We experimented with monochrome images under different types of distortions. Experimental results indicate that the proposed method provides better consistency with the perceived image quality. It is suitable for real applications to control the processed image quality.