Image blind deblurring method based on sparsity metric

A blind deblurring and sparsity technology, applied in the field of image processing, can solve problems such as large difference in effect and insufficient stability, and achieve the effect of good vision and stable image.

Active Publication Date: 2012-11-28
XIDIAN UNIV
View PDF2 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the solution of this method is not stable enough, and the effect of solving different blurred images is v

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image blind deblurring method based on sparsity metric
  • Image blind deblurring method based on sparsity metric
  • Image blind deblurring method based on sparsity metric

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0034] refer to figure 1 , the concrete realization steps of the present invention are as follows:

[0035] Step 1, using the method based on image gradient distribution and the method based on prior characteristics, to obtain two basic blur kernels k f and k s , as the initial blur kernel.

[0036] For methods based on image gradient distribution, see "Removing Camera shake from a Single Photograph" published by Rob Fergus et al., 2006, ACM Transactions on Graphics, vol.25, pp.787–794.

[0037]For methods based on image prior characteristics, see "Blind Deconvolution Using a Normalized Sparsity Measure" published by Dilip Krishman et al., In Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'2011), 2011, pp.233–240 .

[0038] Step 2, using the basic initial blur kernel k f and k s , construct the fuzzy kernel dictionary k d :

[0039] 2a) Initialize the fuzzy kernel k f The linear weighting coefficient of α=0, the blur kernel k ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an image blind deblurring method based on a sparsity metric, and mainly solves the problems in image blind deblurring in the prior art that the noise sensitivity is high and serious ring effect exists. The method provided by the invention comprises the following steps of: (1) obtaining two different fuzzy kernels kf and ks by the existing method; (2) linearly combining the two different fuzzy kernels to obtain a fuzzy dictionary kd={k1, k2...k10}; (3) selecting an image block P with an obvious edge from the fuzzy image, and pretreating the image block P according to a Lucy-Richardson method by virtue of the fuzzy dictionary kd to obtain the pretreated image blocks C1, C2, C3...C10; (4) performing sparsity measurement on the pretreated image blocks C1, C2, C3...C10 to obtain the sparsity S1, S2, S3...S10, and finding out the image block Cmax corresponding to the maximum sparsity Smax; and taking out the fuzzy element kmax corresponding to the image block Cmax; and (5) deblurring the fuzzy image according to the L0-abs algorithm by use of the fuzzy element kmax. In image blind deblurring, the method disclosed by the invention can effectively remove blur and noise and reduce the ring effect as much as possible, and can be applied to the blind deblurring of various fuzzy images.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a blind deblurring method for blurred images, which can be used for deblurring blurred images of various unknown blurring types. Background technique [0002] Image deblurring is an important field in image processing, and its purpose is to restore the original face of the degraded blurred image as much as possible. Image deblurring is divided into two types: non-blind deblurring of images and blind deblurring of images. Image non-blind deblurring refers to the process of obtaining a clear image under the condition of a known degraded blur kernel. This kind of problem has been studied very well, and many existing technologies can obtain a very clear solution. Blind image deblurring refers to the process of estimating the original image from the degraded image without knowing the blur kernel. Blind deblurring of images is very difficult due to less available empirical k...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T5/00
Inventor 王爽焦李成李源梁冲季佩媛王敏郑喆坤
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap