Variance-based no-reference quality assessment of AWGN images

In this paper, a no-reference quality assessment method for image contaminated with additive white Gaussian noise (AWGN) is proposed. The proposed spatial domain method is based on the fact that if the portion of an image having structure is distorted by AWGN, then variance of the distorted image in...

Full description

Saved in:
Bibliographic Details
Published in:Signal, image and video processing image and video processing, 2023-10, Vol.17 (7), p.3575-3583
Main Authors: Baig, Md Amir, Moinuddin, Athar A., Khan, E.
Format: Article
Language:English
Subjects:
Algorithms
Computer Imaging
Computer Science
Contamination
Image Processing and Computer Vision
Image quality
Multimedia Information Systems
Original Paper
Pattern Recognition and Graphics
Quality assessment
Random noise
Signal,Image and Speech Processing
Vision
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, a no-reference quality assessment method for image contaminated with additive white Gaussian noise (AWGN) is proposed. The proposed spatial domain method is based on the fact that if the portion of an image having structure is distorted by AWGN, then variance of the distorted image increases. On the other hand, if the smooth portion of image is contaminated by AWGN, then the distorted image will have proportionate variance as that of AWGN. Therefore, variance of the smoother parts of the image contaminated with AWGN reflects the level of noise and hence can be used as an indicator of the image quality. Extensive simulation results show that the proposed method is highly accurate and lower in complexity in comparison to the existing algorithms.
ISSN:1863-1703
1863-1711
DOI:10.1007/s11760-023-02583-2