Image quality assessment based on improved structural SIMilarity

Authors:
Jinjian Wu;Fei Qi;Guangming Shi
Affiliations:
School of Electronic Engineering, Xidian University, Xi'an, Shaanxi, P.R. China;School of Electronic Engineering, Xidian University, Xi'an, Shaanxi, P.R. China;School of Electronic Engineering, Xidian University, Xi'an, Shaanxi, P.R. China
Venue:
PCM'12 Proceedings of the 13th Pacific-Rim conference on Advances in Multimedia Information Processing
Year:
2012

Citing 6
Cited 1

From Local Kernel to Nonlocal Multiple-Model Image Denoising

International Journal of Computer Vision
Perceptual visual quality metrics: A survey

Journal of Visual Communication and Image Representation
Image quality assessment: from error visibility to structural similarity

IEEE Transactions on Image Processing
Information Content Weighting for Perceptual Image Quality Assessment

IEEE Transactions on Image Processing
Image Quality Assessment by Visual Gradient Similarity

IEEE Transactions on Image Processing
Self-similarity based structural regularity for just noticeable difference estimation

Journal of Visual Communication and Image Representation

Image quality assessment for different wavelet compression techniques in a visual communication framework

Modelling and Simulation in Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a novel image quality assessment (IQA) based on an Improved Structural SIMilarity (ISSIM) which considers the spatial distributions of image structures. The existing structural similarity (SSIM) metric, which measures structure loss based on statistical moments, i.e., the mean and variance, represents mainly the luminance change of pixels rather than describing the spatial distribution. However, the human visual system (HVS) is highly adapted to extract structures with regular spatial distributions. In this paper, we employ a self-similarity based procedure to describe the spatial distribution of image structures. Then, combining with the statistical characters, we improve the structural similarity based quality metric. Furthermore, considering the viewing condition, we extend the ISSIM metric to the multi-scale space. Experimental results demonstrate the proposed IQA metric is more consistent with the human perception than the SSIM metric.