No-reference Image Quality Assessment Based on JND Model and NSCT

DOI

10.2991/mmsta-19.2019.9How to use a DOI?

Keywords

no reference image quality assessment; just noticeable distortion; mutual information; nonsubsampled contourlet transform

Abstract

Considering the image distortion and the masking effect of human visual system, the Just Noticeable Distortion (JND) model is used to construct the model based on the Sigmoid function, using the Just Noticeable Distortion threshold and the absolute value of the pixel change of the image to correct the distorted image and make it conform to the perception effect of the human eye. Then, the mutual information (MI) between relatives' coefficients and father-son coefficients of Nonsubsampled Contourlet Transform(NSCT) directional subbands is calculated as a statistical feature to describe the correlation between these coefficients. Finally, the structural similarity index (SSIM) between relatives coefficients and father-son coefficients of NSCT directional bands is calculated, which is used as a statistical feature to describe the structural information of images. An no-reference image quality assessment (NR-IQA) model is constructed to evaluate image quality. The simulation experiment on LIVE image quality assessment database shows that it can better evaluate the image quality of the distorted image, and accurately capture the edge contour information and texture details of the image, which is better optimized than other algorithms.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xiaosheng Huang
AU  - Feilong Li
AU  - Ruwei Zi
PY  - 2019/12
DA  - 2019/12
TI  - No-reference Image Quality Assessment Based on JND Model and NSCT
BT  - Proceedings of the 2019 2nd International Conference on Mathematics, Modeling and Simulation Technologies and Applications (MMSTA 2019)
PB  - Atlantis Press
SP  - 39
EP  - 43
SN  - 2352-538X
UR  - https://doi.org/10.2991/mmsta-19.2019.9
DO  - 10.2991/mmsta-19.2019.9
ID  - Huang2019/12
ER  -