Related Experiment Video
Updated: Apr 23, 2026
07:45
Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
Published on: July 21, 2020
4.0K
Blind image blur assessment using singular value similarity and blur comparisons
Qing-Bing Sang1, Xiao-Jun Wu1, Chao-Feng Li1
1Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education), School of Internet of Things Engineering, Jiangnan University, Wuxi, Jiangsu, China.
Plos One
|September 24, 2014
Summary
This study introduces a novel blind blur index for objective image quality assessment (IQA). The new method uses singular value similarity to accurately predict human judgment of image blur and noise.
Area of Science:
- Computer Vision
- Image Processing
- Signal Processing
Background:
- Growing demand for high-quality consumer images necessitates robust objective image quality assessment (IQA).
- Existing no-reference IQA algorithms face challenges in accurately evaluating image blur and noise without a reference image.
Purpose of the Study:
- To propose a novel blind blur index for still images.
- To develop an algorithm for no-reference objective image quality assessment (IQA) based on singular value similarity.
Main Methods:
- A re-blurred image is generated using Gaussian blur on the test image.
- Singular value decomposition (SVD) is applied to both the test and re-blurred images.
- An image blur index is computed based on the similarity of singular values.
Main Results:
- The proposed algorithm demonstrates a high correlation with human judgment.
- Experimental results validate the algorithm's effectiveness on simulated databases for assessing blur and noise distortion.
Conclusions:
- The developed blind blur index offers a reliable method for no-reference IQA.
- Singular value similarity provides a promising approach for objective image quality evaluation, particularly for blur and noise.
