Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Single image blind deblurring method based on MAP method

A blind deblurring, single-image technology, applied in image enhancement, image data processing, instruments, etc., can solve problems such as large limitations and weak generalization ability of related algorithms, and achieve the effect of rapid convergence

Inactive Publication Date: 2019-06-07
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The problem to be solved by the present invention is that there are some deficiencies in the current correlation blind single image restoration method, for example, the correlation algorithm has weak generalization ability and particularly large limitations, and cannot be well used in other scenarios

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
  • Single image blind deblurring method based on MAP method
  • Single image blind deblurring method based on MAP method
  • Single image blind deblurring method based on MAP method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Example 1: Such as Figure 1-4 As shown, the experimental data of this embodiment includes 23 blurred images, in which there are low-illuminance images, human faces and natural images, and the blur kernels are all generated by the same method. The parameters are set as follows: α = 0.005, β = 0.0005, γ = 0.004, μ max = 2 3 , And ω max = 2 2 . Take one of the images as an example.

[0038] Specific steps are as follows:

[0039] Step1: First convert the image to be restored into a grayscale image, and then convert the obtained grayscale image into a double type image y to facilitate subsequent calculations.

[0040] Step2: Initialize the relevant parameters (k, x).

[0041] Step3: Solve u, g, d by x.

[0042] Step3.1 Substitute x into the equation: , And initialize μ←2α (assignment operation is only performed when initial value is assigned) to get the value of u.

[0043] Step3.1 Substitute x into the equation: And initialize (Assignment operation is only performed when init...

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 relates to a single image blind deblurring method based on an MAP method, and belongs to the field of digital image restoration. The method comprises the following steps: converting a color image into a grayscale image; adding the priori knowledge such as dark channel, intensity and gradient priori to an objective function; solving a non-convex problem in the objective function by adopting a semi-quadratic separation method; and for the obtained blurred kernel, restoring a clear image by adopting a non-blind method. According to the method, the blind deblurring problem on the natural image is effectively solved by combining dark channel priori knowledge, intensity priori knowledge and gradient priori knowledge, and for the natural image, the dark channel priori knowledge hasa very good sparse effect. And for some edges of the natural image, the intensity priori can well retain the edges. Gradient prior can well restrain the artifact problem in the deblurring process. Therefore, the natural image can be well restored by combining the priori so as to obtain a better clear image.

Description

Technical field [0001] The invention relates to a blind deblurring method for a single image based on a MAP method, which belongs to the field of digital image restoration. Background technique [0002] Blind image deblurring is a classic problem in the field of image processing and computer vision. Its goal is to restore the hidden image in the blurred image. When the fuzzy shape satisfies the spatial invariance, the fuzzy process can be modeled in the following ways: [0003] y=x*k+n (15) [0004] Where * represents the convolution operator, y, x, k and n represent blurred images, clean images, blur kernels and noise, respectively. The problem in the above formula is ill-posed, because the kernel k is all unknown and there are infinitely many solutions. In order to solve this problem, additional constraints and prior knowledge about the blur kernel and the image are necessary. [0005] The success of recent deblurring methods mainly comes from the research progress of effective i...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T5/00
Inventor 许煜刘辉尚振宏李润鑫罗静
Owner KUNMING UNIV OF SCI & TECH
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
Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com