A Variational PDE Image Restoration Method with Boundary and Structure Preservation

An image and degraded image technology, applied in the field of image restoration, can solve problems such as poor structural features, unfavorable processing, and low computational efficiency

Inactive Publication Date: 2017-12-19
HENAN UNIVERSITY
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to overcome the disadvantages of the above-mentioned technologies, such as poor structural features such as maintaining image corners and edges and smooth gradient regions, low computational efficiency, and unfavorable processing of large data, and provide a variational PDE image edge-preserving Structural restoration method, which combines the fully variable fractal functional space and adaptive parameter selection criteria, adopts the strategy of transforming the objective function of the model into a strategy that can efficiently solve multiple sub-problems, establishes a new restoration model and develops a stable and efficient Apply software to restore high-quality images, so as to provide strong theoretical support and technology for image segmentation, fusion, completion, registration and recognition in remote sensing image processing, medical image processing, video target tracking and image structure feature extraction support

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
  • A Variational PDE Image Restoration Method with Boundary and Structure Preservation
  • A Variational PDE Image Restoration Method with Boundary and Structure Preservation
  • A Variational PDE Image Restoration Method with Boundary and Structure Preservation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0075] Example 1 Image denoising experiment:

[0076] Such as figure 2 As shown, through comparison, it is not difficult to find that the invented model can obviously maintain the structural features of the image, such as the edge area and smooth gradient area of ​​the image. The SNR of the restored image of the ROF model is 17.7677, the SNR of the high-order model is 17.9998, and the SNR of the invented model is 18.3694.

Embodiment 2

[0077] Example 2 Image deblurring experiment:

[0078] Such as image 3 As shown, the invented model can better restore the degraded image, and is more effective in maintaining the detailed features and structural features of the image. The signal-to-noise ratio of the restored image of the ROF model is 14.3778, the signal-to-noise ratio of the high-order model is 14.4778, and the signal-to-noise ratio of the invented model is 14.5118.

Embodiment 3

[0079] Example 3 Image deblurring experiment:

[0080] Such as Figure 4 As shown, it can be seen from the partially enlarged image that the ROF model contains obvious letter information in the stem of the pepper. The high-level model causes the boundary of the restored image to be blurred. However, the proposed model can well overcome the defects of these two types of models. Correspondingly, the signal-to-noise ratio of the restored image of the ROF model is 15.4055, the signal-to-noise ratio of the high-order model is 16.4336, and the signal-to-noise ratio of the invented model is 16.6034.

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 variational PDE image edge-preserving structure restoration method, which introduces a structural feature detection function into the model to punish different functional spaces in the model, thereby making the model self-adaptive and obtaining a high-quality restored image . In the process of implementing the model algorithm, we first transform the problem into multiple sub-problems that are easy to solve, and use the periodic difference format, so that the numerical model has the characteristic of using fast Fourier changes, and can be solved efficiently and stably. The software package in the present invention focuses on transforming the model into sub-problems that can be solved quickly. Since the sub-problems are separable, and the fast Fourier transform can be used by using the cyclic difference format, it can be solved efficiently and stably and has parallelism sex.

Description

Technical field [0001] The invention belongs to the technical field of image restoration technology, and relates to a method for restoring a variational PDE image with edge preserving structure. Background technique [0002] In the process of image formation, recording, processing, and transmission, due to imperfect imaging systems, recording equipment and transmission media, noise, blur, or partial information loss and other quality degradation problems are often introduced into the image. For example, the disturbance effect of atmospheric turbulence, the noise interference caused by changes in environmental conditions and the quality of the sensor components themselves, and the motion blur caused by the relative movement between the subject and the imaging device. Image restoration is a technology that uses some prior knowledge of the image to reconstruct the image to improve the image quality. However, as the requirements for image quality and image resolution increase, the c...

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 Patents(China)
IPC IPC(8): G06T5/00
Inventor 庞志峰宋瑞丽张彦周张培彦李晓燕蒋妍
Owner HENAN UNIVERSITY
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