NUMERICAL METHODS AND DATA ANALYSIS
Image blur estimation using gradient field analysis
D. G. Asatryanab
a Russian-Armenian (Slavonic) University, Yerevan, Armenia
b Institute for Information and Automation Problems of National Academy of Sciences of Armenia, Yerevan, Armenia
Estimating the degree of blur is an important step in improving the image quality. In the literature, many approaches, criteria and algorithms for estimating the degree of blurring are proposed, which utilize the properties of the gradient field of an image. In this paper, we propose a new measure for blur estimation, based on the use of the Weibull distribution shape parameter, determined from a sample of magnitudes of the image gradient. Using artificially blurred images as an example, it is shown that the larger the blur factor, the nearer the proposed measure value to "2", and a monotonic dependence of the measure value on the blur factor is observed. The same effect is observed when the image is filtered, but as the filter factor increases, the value of the measure decreases monotonically. In the paper, it is proposed that the measure of blurring should be considered as a criterion for the structuredness of the image.
image blur, gradient magnitude, Weibull distribution, form parameter, Sobel operator, structure.
PDF file (2010 kB)
D. G. Asatryan, “Image blur estimation using gradient field analysis”, Computer Optics, 41:6 (2017), 957–962
Citation in format AMSBIB
\paper Image blur estimation using gradient field analysis
\jour Computer Optics
Citing articles on Google Scholar:
Related articles on Google Scholar:
|Number of views:|