Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Fuzzy kernel computation method for motion blurred image restoration

A motion blurred image and calculation method technology, applied in the field of blur kernel calculation, can solve the problems of long time consumption, unsatisfactory deblurring effect, slow calculation and operation of blurring kernel parameters, etc., and achieve short time consumption, fast calculation speed and deblurring effect Good results

Active Publication Date: 2018-12-14
SOUTHEAST UNIV
View PDF4 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Purpose of the invention: The purpose of the present invention is to provide a blur kernel calculation method for motion blurred image restoration, to solve the problem of unsatisfactory deblurring effect, slow calculation of blur kernel parameters, and long time-consuming problems

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
  • Fuzzy kernel computation method for motion blurred image restoration
  • Fuzzy kernel computation method for motion blurred image restoration
  • Fuzzy kernel computation method for motion blurred image restoration

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The present invention will be further described below in conjunction with the accompanying drawings.

[0020] Such as Figure 1-2 As shown, a blur kernel parameter estimation method for motion blur image restoration, including the following steps:

[0021] (1) Establish an image degradation model. The key to image deblurring is to establish an image degradation model. For digital images, the formation of motion blurred images can usually be described as the convolution process of the original clear image I(x,y) and the blur kernel h(x,y). If the noise exists, the influence of noise should be considered, generally assuming that the noise is additive noise, the process can be expressed as:

[0022] g(x,y)=h(x,y)*I(x,y)+n(x,y) (Formula 1)

[0023] Among them, * indicates convolution operation, n(x,y) indicates additive noise, g(x,y) indicates blurred image, I(x,y) indicates clear image, and h(x,y) indicates blurred image The blur kernel determined by the two parameters...

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 a fuzzy kernel computation method for motion blurred image restoration. The method is a fuzzy kernel parameter estimation algorithm based on sparse characteristic, super Laplace a priori and integrated BP neural network. Firstly, under the constraint condition that the image gray gradient conforms to super Laplace distribution, the fuzzy angle of the blurred image is determined by analyzing the sparse representation coefficient of the blurred image. Then, the amplitude sum of Fourier coefficients obtained from Fourier transform is used as input to estimate the fuzzy length by training the integrated BP neural network model based on Bagging method. Finally, the deblurring image is obtained by one-step deblurring algorithm with known blurring kernel. The invention hasthe advantages of accurate estimation of fuzzy kernel parameters, fast operation speed, short time consuming and good deblurring effect, and can make the edge of the recovered image clearer and the ringing effect less by recovering the motion blurred image.

Description

technical field [0001] The invention relates to image signal processing, in particular to a blur kernel calculation method for motion blur image restoration. Background technique [0002] Blurred images are usually caused by relative camera or object motion, atmospheric turbulence, camera out of focus, and data loss during data transmission. Deblurring of motion blurred images is one of the important topics in image restoration. According to whether the blur kernel is known, image deblurring can be divided into non-blind and blind deblurring. When the blur kernel is known, it is called non-blind deblurring. From a mathematical point of view, the main purpose of non-blind image deblurring is to implement a deconvolution and add appropriate constraints in the solution process. At present, non-blind image deblurring technology is quite mature and has been successfully applied in many scientific research fields. On the other hand, blind image deblurring techniques need to au...

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): G06T5/00G06N3/08
CPCG06N3/084G06T2207/20084G06T2207/20081G06T5/73
Inventor 陈熙源柳笛方文辉刘晓
Owner SOUTHEAST UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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