Paper Info
Title
Motion deblurring from a single image using gradient enhancement
Abstract
Motion deblurring is one of the recovery problems in image restoration, which remains several challenges in kernel estimation and blind deconvolution. This paper proposes a new optimization method for estimating the blurring kernel by gradient enhancement, which can iteratively solve a uniform deblur model. In this model, the point-spread-function(PSF) can be accurately estimated and refined by gradually enhancing the image gradients. Our approach includes following steps: edge-preserving gradient enhancement, edge selection, kernel estimation and refinement, fast non-blind deconvolution. The edge-preserving gradient enhancement can restore sharp edges while have no effect in flat regions. Combined with the edge selection, it greatly helps to estimate the kernel. To improve its speed performance, the estimation and deconvolution steps are executed in frequency domain. Experimental results demonstrate that our method can efficiently produce an accurate blur kernel and a restored image with fine image details.
Year
DOI
Venue
2011
10.1145/2087756.2087800
VRCAI
Keywords
Field
DocType
deconvolution step,edge selection,fine image detail,accurate blur kernel,image gradient,gradient enhancement,kernel estimation,edge-preserving gradient enhancement,image restoration,single image,blind deconvolution,frequency domain,point spread function
Kernel (linear algebra),Frequency domain,Computer vision,Deblurring,Blind deconvolution,Computer science,Deconvolution,Artificial intelligence,Image restoration,Kernel density estimation
Conference
Citations 
PageRank 
References 
0
0.34
17
Authors
4
Name
Order
Citations
PageRank
Jiahua Chen100.34
Zhifeng Xie25310.70
Bin Sheng336861.19
Lizhuang Ma4498100.70