De-noising method integrated with gradient histogram and low rank constraint

A gradient histogram, low-rank constraint technology, applied in image enhancement, image data processing, instruments, etc., can solve the problem of poor denoising effect of denoising methods, and achieve the effect of enhancing denoising performance

Active Publication Date: 2017-11-17
GUILIN UNIV OF ELECTRONIC TECH
View PDF9 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is that the existing denoising method has

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
  • De-noising method integrated with gradient histogram and low rank constraint
  • De-noising method integrated with gradient histogram and low rank constraint
  • De-noising method integrated with gradient histogram and low rank constraint

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] The technical solutions of the present invention will be described in detail below in conjunction with the accompanying drawings in specific embodiments of the present invention.

[0044] A denoising method that combines gradient histograms with low-rank constraints, such as figure 1 As shown, it specifically includes the following steps:

[0045] Step 1, read the noisy image, and obtain the matrix expression y of the image;

[0046] Step 2, logarithmically transform the noisy image to obtain the noisy image f in the logarithmic domain, and initialize the number of iterations k=1, z (0) =f, where z (0) is the image to be restored in the logarithmic domain, is the gradient of z;

[0047] Step 3, use the sliding window technique to convert the noisy image z (k-1) Divide into small blocks of 7×7, and use K-means to cluster these small image blocks into 70 categories;

[0048] Step 4, use the principal component analysis method to calculate the sub-dictionary in ea...

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 de-noising method integrated with a gradient histogram and low rank constraint. Based on sparse priori and other non-local self similar priori, sparse expression advantages are utilized, and a non-local regular item, a gradient regular item and a low rank constraint item are added to remove multiplicative noise. The method is advantaged in that a multiplicative noise model is converted through logarithmic transformation into an additive noise model of a logarithm domain, a training dictionary of the noise image in the logarithm domain is utilized, image gradient histogram estimation and low rank constraint are combined, local and non-local image connection is enhanced, on the condition that de-noising is effectively realized, the image texture information is better reserved; the experiment result is relatively good in subjective vision and objective evaluation indexes, the precise texture structure of the image is reserved to a great degree, and the image after de-noising is more intelligible.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to a denoising method combining gradient histograms and low rank constraints. Background technique [0002] Image noise removal is one of the most basic problems in digital image processing. Images are more or less inevitably polluted by random noise. The mathematical model of multiplicative noise is: y=x v, where y is a noisy image, and x represents Original image, v is noise. Sparse priors and non-local self-similar priors are widely used in image denoising. In addition, the joint use of sparse priors and other non-local self-similar priors has also resulted in many advanced image denoising algorithms. Such as using curvelet transform as its structure and using l 1 Image denoising algorithm with norm acting on sparsely coded coefficients (DFN model), image denoising algorithm by sparse regularization variation with MAP estimation and sparse represen...

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
CPCG06T5/002
Inventor 陈利霞李佳宇王学文何成凤
Owner GUILIN UNIV OF ELECTRONIC TECH
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