Content uploaded by Hongjun Li
Author content
All content in this area was uploaded by Hongjun Li on Nov 19, 2015
Content may be subject to copyright.
A preview of the PDF is not available
No-reference Image Quality Assessment Based on Parameters of wavelet Coeffieients Distribution
Abstract
We propose a no-reference method for image quality assessment to be used to various types of image distortions. We assess the image using statistic information of natural scenes, and use the generalized Gaussian density model to fit the marginal distribution of wavelet coefficients. Degree of image distortion is measured with the parameter values in the generalized Gaussian density model. Objective assessment of the image quality is obtained by quantifying the difference between the values of scale and shape parameters. Experimental results are consistent with subjective assessments, showing that the proposed method can be applied to most common types of image distortion to give good prediction.
ResearchGate has not been able to resolve any citations for this publication.
A new no-reference image quality assessment method was proposed based on the statistical model of natural images in order to assess the distortion of white noise, Gaussian blur and JPEG2000 compression. For natural images, the means of Contourlet subband coefficient amplitudes (MSC) decreased approximately linearly with scale index. Then MSC of high frequency was predicted from MSC of low frequency, which was not badly affected by distortion. The distortion was evaluated using the difference between the predicted means and the real means in distorted image. The metric of subband distortion for different scales, different directions, and different regions was weighted and pooled into a quality score considering characteristic of human visual system. Experimental results showed that the method was consistent with subjective assessment, and exceeded the performance of peak signal to noise rate (PSNR).
Presuming that human visual perception is highly sensitive to the structural information in a scene, we propose the concept of structural activity (SA) together with a model of SA indicator in a new framework for no-reference (NR) image quality assessment (QA) in this study. The proposed framework estimates image quality based on the quantification of the SA information of different visual significance. We propose some alternative implementations of SA indicator in this paper as examples to demonstrate the effectiveness of the SA-motivated framework. Comprehensive testing demonstrates that the model of SA indicator exhibits satisfactory performance in comparison with subjective quality scores as well as representative full-reference (FR) image quality measures.