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

Image restoration method based on Criminisi

An image and picture technology, applied in the field of image restoration based on Criminisi, which can solve the problems of algorithm stop, texture extension overflow, image texture information repair error, etc.

Active Publication Date: 2020-05-19
ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The present invention improves the calculation formula of the priority in combination with the least square fitting, and can solve the problem that the algorithm stops when the priority is zero; the phenomenon of texture extension and overflow in the Criminisi algorithm and the problem of repairing errors when the image texture information is strong

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 restoration method based on Criminisi
  • Image restoration method based on Criminisi
  • Image restoration method based on Criminisi

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] Such as figure 1 As shown, I is the image to be repaired, Ω is the area to be repaired, is the edge of the area to be repaired, that is, the critical area between the area to be repaired and the known area, φ is the source area, that is, the known area of ​​the image to be repaired, φ=I-Ω. P is the target pixel, Ψp is the block to be repaired centered on p, n p is the normal direction of p, that is, the unit vector perpendicular to the edge of the area to be repaired at point P, is the direction of the isolux line of p, ▽I p is the tangent direction of the iso-illuminance line of p, and is also the direction of the gradient.

[0050] The Criminisi algorithm can be divided into three steps: calculating the priority, finding the best matching block, and updating the confidence.

[0051] First calculate the priority, the formula for calculating the priority is:

[0052] P(p)=α·C(p)+β·D(p) (1)

[0053] Among them, α and β are weighting coefficients, and the least squ...

example 1

[0094] Example 1: Applied to the roof restoration comparison chart

[0095] image 3 The experiment selected a roof image, which reflects the algorithm's processing of simple texture images with a single color, from figure 2 It can be seen from (b) that the Criminisi algorithm can handle the structure of the upper and lower parts of the house better, but there is a big gap in the processing of the middle texture part. The improved algorithm such as figure 2 As shown in (c), not only can the roof structure be completely processed, but also the details of the texture are well restored.

[0096] The mathematical parameter confidence α and data item β in the algorithm are calculated by the least squares method through the pixels of the sound image and the coding image of the experimental image in the embodiment. When the one-dimensional array (x1, y1), (x2, y2), ..., (xn, yn) is (0.53, 1.2506), (0.751, 1.4804), ..., (1.327, 2.044), enter the formula (11 ) and (12) get

[00...

example 2

[0101] Example 2: Applied to building restoration comparison chart

[0102] Figure 4 The experiment selects a building map, which reflects the treatment of structural integrity and extensibility by the Criminisi algorithm, from image 3 It can be seen from (c) that the structure of the picture is severely missing, the upper half of the beam is only half repaired, and the shape of the drop-shaped structure in the lower half is also deformed. An improved algorithm such as image 3 As shown in (c), the beam in the upper half has been completely repaired, and the drop-shaped structure in the lower half has been basically restored.

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 restoration method based on Criminisi. The method comprises the following steps: 1, reading a to-be-restored image I and determining a to-be-restored area omega; 2, selecting a point p with the highest priority from the edge of a to-be-repaired area, constructing a block psi p needing to be repaired currently with p as the midpoint, wherein the edge of the to-be-repaired area is the critical value of the to-be-repaired area and a known area; 3, searching an optimal matching block psi q in a source region phi, and copying pixel information of the psi q to psi p,the source region phi being a known region of the to-be-restored image, and phi = I-omega; and step 4, updating the edge of the to-be-restored area until omega is 0, and completing image restoration.In conclusion, the method has the characteristics of high practicability and obvious restoration effect, and can be applied to the fields of medical images, public security case detection, movie andtelevision production, aerial images, cultural relic protection and the like.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and more specifically relates to a Criminisi-based image restoration method. Background technique [0002] Digital image restoration technology is a hot topic in the field of digital image processing. It is also playing an increasingly important role in modern society and involves all aspects of social life. It is mainly used in medical imaging, public security case detection, film and television production, and aerial photography. Images, protection of ancient cultural relics and other fields. Image restoration mainly restores the missing or damaged part of the image through the known information of the image. Image restoration methods can be divided into two categories: image restoration methods based on partial differential equations (Partial Differential Equation, PDE) and texture-based methods. Synthetic Image Restoration Methods. The image restoration method based on PDE ...

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): G06T5/00G06T5/20
CPCG06T5/20G06T2207/10004G06T5/00
Inventor 陈燕王凤琴李素萍
Owner ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY
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