Edge-based multi-direction weighting TV and self-similarity constraint image defuzzification method

A self-similarity and deblurring technology, applied in image enhancement, image data processing, instruments, etc., can solve problems such as insufficient image prior information expression ability, limited ability to restore details and texture, and error of spatial domain weight matrix

Inactive Publication Date: 2017-03-08
TIANJIN UNIV
View PDF1 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among many deblurring methods, total variation (TV) regularization is widely used in image denoising, image deblurring, etc. [1,2] , but its detail and texture recovery capabilities are limited
[0003] On the other hand, the traditional TV model only considers the local features of the image, and does not utilize the structurally similar global image block information in the image, that is, the non-local self-similarity of the image [1] , the ability to express the prior information of the image is insufficient
Existing methods incorporate spatial non-local self-similarity into deblurring models [5] , but the airspace weight matrix is ​​easy to introduce a large error

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
  • Edge-based multi-direction weighting TV and self-similarity constraint image defuzzification method
  • Edge-based multi-direction weighting TV and self-similarity constraint image defuzzification method
  • Edge-based multi-direction weighting TV and self-similarity constraint image defuzzification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0099] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the described specific embodiments are only for explaining and illustrating the present invention, and are not intended to limit the present invention.

[0100] Traditional weighted TV (Weighted Total Variation, WTV) model:

[0101]

[0102] Among them, the weight g i is defined as

[0103]

[0104] In the formula, x is the original image, and f is the blurred image. Among them, k is a threshold parameter, which is used to judge whether the current pixel belongs to the smooth area or the edge structure.

[0105] Where the gradient is large, the weight g i Take a smaller value to reduce the smoothness and better protect the edge; where the gradient is small, the weight g i Take a larger value to achieve strong smoothing and remove noise. However, the traditional weighted TV only determines the wei...

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 edge-based multi-direction weighting TV and self-similarity constraint image defuzzification method. A conventional TV model is improved based on edge detection, and transform domain self-similarity of an image is combined. Firstly, pixels in a center pixel neighborhood is divided into same-side pixel pairs and different-side pixel pairs by edge detection, different weights are adopted for different types of the pixels to obtain a multi-direction weighting TV algorithm, and the detailed information of the image is reserved to the greatest extent in defuzzification; secondly, a transform domain self-similarity regularization term is fused into the TV model, so that the limitation that the conventional non-local regularization depends on a non-local weight matrix is overcome; therefore, the texture and structural self-similarity of the image can be described more accurately; and the defuzzification visual effect of the image is further improved. By adoption of the defuzzification method, the visual effect is improved by defuzzification while the edge information and important details of the image are well reserved; and in addition, the objective indicators, such as a peak signal to noise ratio and the like, of the image are improved.

Description

technical field [0001] The invention belongs to the field of computer image processing, and is mostly used in related fields such as image or video deblurring. Background technique [0002] Image deblurring has always been a hot spot in the field of computer vision and image processing, and has attracted much attention because of its cutting-edge and wide application. Among many deblurring methods, total variation (TV) regularization is widely used in image denoising, image deblurring, etc. [1,2] , but its detail and texture recovery capabilities are limited. [0003] On the other hand, the traditional TV model only considers the local features of the image, and does not utilize the structurally similar global image block information in the image, that is, the non-local self-similarity of the image [1] , the ability to express the prior information of the image is insufficient. Existing methods incorporate spatial non-local self-similarity into deblurring models [5] , bu...

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
CPCG06T5/003
Inventor 杨爱萍张越王建田玉针
Owner TIANJIN 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