Noise blurred image blind deconvolution method based on image significant structure

A blurred image and deconvolution technology, applied in image enhancement, image data processing, instruments, etc.

Inactive Publication Date: 2017-07-18
SHENYANG INST OF AUTOMATION - CHINESE ACAD OF SCI
View PDF2 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the shortcomings of the current single image blind convolution technology in processing noise-blurred images, the present invention provides a noise-blurred

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
  • Noise blurred image blind deconvolution method based on image significant structure
  • Noise blurred image blind deconvolution method based on image significant structure
  • Noise blurred image blind deconvolution method based on image significant structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0046] The present invention first suppresses the image noise through noise reduction preprocessing, uses the method based on the total variation model to extract the salient structure of the blurred image, and then uses the gradient selection method to remove the salient edges that are not conducive to the estimation of the blur kernel, thereby improving the estimation of the blur kernel. The robustness of the blur kernel; then a two-stage blur kernel estimation strategy is adopted, and the blur kernel estimation method based on the salient structure of the image and ISD technology are used to realize the accurate estimation of the blur kernel; finally, the final non-blind image deconvolution method with sparse prior constraints is used. image restoration.

[0047] exist figure 1 In this paper, after the BM3D filter is used to denoise and sm...

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 noise blurred image blind deconvolution method based on an image significant structure. The noise blurred image blind deconvolution method comprises steps that to-be-deblurred image data is input; the denoising pre-treatment of the input image data is carried out; the significant edge extraction of the image after the denoising pre-treatment is carried out; an image strong edge is reconstructed by carrying out the Shock filtering of the image after the significant edge extraction; the image significant edge used for blurred kernel estimation is calculated by using the image strong edge; an initial blurred kernel is estimated; the rough image restoration is carried out by using the estimated initial blurred kernel; the blurred kernel based on ISD after the rough image restoration is corrected; and the image is restored. The problem of the image deblurring sensitive to noises is solved, and the blurred kernel of the noise blurred image is estimated accurately, and high quality restored image is provided.

Description

technical field [0001] The invention relates to the fields of computer vision and digital image processing, in particular to a blind deconvolution method for noise-blurred images based on image salient structures. Background technique [0002] Blind deconvolution of a single image is one of the most basic research issues in the field of image processing and computer vision. field. The problem of blind deconvolution of a single image is essentially an ill-conditioned inverse problem in mathematics. It is usually solved by estimating the blur kernel first, converting the deconvolution problem into a linear image restoration problem, and then restoring the clear image through a non-blind deconvolution method, which can achieve better results. During non-blind deconvolution, the accuracy of blur kernel estimation directly affects the quality of image restoration. Therefore, for the problem of blind deconvolution of a single image, it is crucial to find a relatively accurate e...

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 赵怀慈孙士洁吕进锋郝明国李波
Owner SHENYANG INST OF AUTOMATION - CHINESE ACAD OF SCI
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