Impact
Impact
of Directional Variograms in Fuzzy Type-II Punctual Kriging based Image
Restoration
Asmatullah Chaudhry1, 2,
Jin Young Kim2, Tran Anh Tuan2
1. HRD, PINSTECH, P.O. Nilore, Islamabad, Pakistan
2. School of Electronics & Computer
Engineering, Chonnam National University (CNU), Gwangju, South Korea
[email protected], [email protected]
Abstract: In
this paper, we perform an experimental study to investigate directional
variograms in punctual kriging and consequently its effect on image
restoration. We employ punctual kriging in conjunction with fuzzy logic type-II
and fuzzy smoothing based approaches to remove white Gaussian noise from
corrupted images. Images degraded with Gaussian white noise are restored by
first utilizing fuzzy logic for selecting pixels that needs kriging. Type-II
fuzzy set has been used to generate fuzzy map for detection of noisy pixels.
Further, local neighborhood information is used to ensure noise free pixels in
3x3 window to estimate the noisy data. The concept of directional semivariance
based punctual kriging is then used to estimate the intensity of a noisy pixel.
Image restoration performance based comparison has been made against adaptive
Weiner filter and existing fuzzy non-directional kriging approaches.
Experimental results on various images and different image quality measures
show that directional semivariance may provide information how image data is
distributed in different directions but could not play an effective role in
punctual kriging based image restoration as compared to non-directional
punctual kriging estimation.
[Chaudhry
A, Kim JY. Impact of Directional Variograms in Fuzzy Type-II Punctual
Kriging based Image Restoration. Life
Sci J 2013;10(1):1010-1021] (ISSN:1097-8135). http://www.lifesciencesite.com.
Keywords: Image restoration; Type-II fuzzy logic; punctual kriging; directional variograms; adaptive spatial filtering
