Inconsistent image blind restoration method based on sparse representation

A sparse representation and blind restoration technology, applied in the field of image restoration, which can solve the problems of insufficient punishment in flat areas, difficult norm constraints, and prone to ringing.

Inactive Publication Date: 2016-09-21
TIANJIN UNIV
View PDF7 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the complexity of natural images, the blur kernel estimated from a single image according to this model is likely to have large errors in some areas
Natural image gradient heavy-tailed distribution features [2] and fuzzy kernel sparse normalization [3] The regularization method as a priori knowledge is a typical method to solve the above problems, but this method does not punish the flat area enough, and the norm constraint is difficult to completely approximate the heavy-tailed distribution of the gradient, which is prone to ringing and the restoration effect is not ideal

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
  • Inconsistent image blind restoration method based on sparse representation
  • Inconsistent image blind restoration method based on sparse representation
  • Inconsistent image blind restoration method based on sparse representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] Taking the "Buddha statue" in Fig. 2(a) as an example, a non-uniform image blind restoration method based on sparse representation proposed by the present invention is used for restoration, so as to obtain the original clear image after restoration, as shown in Fig. 2(e). like figure 1 As shown, its processing process includes the following steps:

[0066] Step 1. Build an image blind restoration model, including:

[0067] According to the three-dimensional camera shake model, the non-uniform image blur degradation model shown in formula (1) is constructed:

[0068] y = Σ k w k ( Σ j C i j k x j ) + ϵ - - - ...

Embodiment 2

[0099] Similarly, taking the "fish" in Figure 3(a) as an example, a non-uniform image blind restoration method based on sparse representation proposed by the present invention is used for restoration, so as to obtain the original clear image after restoration, as shown in Figure 3(e) shown. Figure 3(b) to Figure 3(d) is the restoration result using prior art methods. It can be seen that the restoration result using the Levin algorithm shown in Figure 3(b) is obviously distorted and the dorsal fin is deformed, the restoration result using the Babacan algorithm shown in Figure 3(c) is better, and Figure 3(d) shows The restoration result using the Hu algorithm is obviously noisy and the details are lost, but the restoration result of the algorithm of the present invention is better as shown in Figure 3(e), and the texture of the fish fin and rope cloth can be seen, and the overall effect is natural. In a word, the algorithm of the present invention has obvious advantages in res...

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 inconsistent image blind restoration method based on sparse representation. The method comprises steps of: creating an inconsistent image fuzzy degeneration model depending on a camera three-dimensional shaking model in combination with over-complete dictionary representation of a natural image; inputting a fuzzy image to be restored and an over-complete dictionary to solve an initial sparse coefficient and initializing a parameter; using the over-complete dictionary representation of the natural image sparsity of the fuzzy core and sparse coefficient as the regular constraint of the model, and transforming the resolution of the inconsistent blind image restoration model into multiple simple subproblems by using an alternate iteration method so as to achieve blind restoration of the fuzzy image y. The method has better restoration effect on the fuzzy image acquired on natural condition, achieves restored images with clear details, no distortion, and low noise, has better visual effects and extendibility.

Description

technical field [0001] The invention relates to a computer image processing method, in particular to an image restoration method. Background technique [0002] When the imaging equipment is collecting images, artificial shaking or inherent mechanical shaking of the equipment will cause the collected images to appear overall blurred, the boundary of the target object is not clear, and the details of the information are lost. application poses a major problem. Therefore, blurred image restoration has become a hot problem to be solved urgently. The existing blurred image blind restoration can be divided into consistent image blind restoration and non-uniform image blind restoration according to whether the blur kernel is globally consistent. In fact, the shaking of the camera in three-dimensional space will cause non-uniform blur in the imaging plane. Therefore, blind restoration of non-uniform images is more practical. [0003] By studying the spatial geometric model of cam...

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
CPCG06T5/92G06T5/73
Inventor 杨爱萍王南梁斌何宇清魏宝强
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