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

Image denoising method based on mixed robust weights and method noise (MN)

A method of noise and image technology, applied in the field of intelligent information processing, can solve the problem that the NLM algorithm cannot effectively balance the relationship between noise suppression and detail preservation, and achieve the effect of improving accuracy

Active Publication Date: 2018-09-28
JIANGNAN UNIV
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0017] The present invention aims to propose an image denoising method based on hybrid robust weight and method noise to solve the problem that the traditional NLM algorithm cannot effectively balance the relationship between noise suppression and detail preservation, and effectively preserve the image while suppressing noise. structural details of

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 denoising method based on mixed robust weights and method noise (MN)
  • Image denoising method based on mixed robust weights and method noise (MN)
  • Image denoising method based on mixed robust weights and method noise (MN)

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] Embodiment 1: An image denoising method based on mixed robust weights and method noise, refer to figure 1 ;

[0064] Step 1. Input a noisy image Y={y(i)|i=1,2,...,n}, where y(i) is the gray value of pixel i, and n is the total number of pixels. The noise-containing image Y is pre-denoised by bilateral filtering (BF) algorithm, and the pre-denoising image D is obtained. pre ={d pre (i)|i=1,2,...,n}:

[0065]

[0066]

[0067] Where i, j are the i-th pixel and j-th pixel in the image respectively, N i Represents a neighborhood block centered on pixel i (the size is 5×5), dist(i, j) is the Euclidean distance between pixel i and j, and the parameter σ S and σ G Represent the spatial proximity coefficient and the gray similarity coefficient (the values ​​are respectively taken as 3.0 and 0.7σ, σ is the noise standard deviation of the image), and ω(i) is the normalization item;

[0068] Step 2. Use the non-local mean denoising (HRW-NLM) method based on hybrid robu...

Embodiment 2

[0081] Embodiment 2: effect simulation experiment of the present invention

[0082] Four standard images Lena (512×512), Barbara (512×512), Peppers (256×256), House (256×256) are used for testing, and the mean value is 0 and the standard deviation σ is 15 and 25 respectively. , 35, 50, 80, 100 Gaussian white noise. The simulation experiment is mainly carried out from the following two aspects: the performance comparison of the three NLM algorithms (NLM, HRW-NLM and the inventive method HRWIMN-TSNLM) proposed in the present invention and the comparison between the inventive method (HRWIMN-TSNLM) and other several algorithms ( NLM, MN-TSNLM, BFNLM, MNLM) comparison. The experimental testing environment is MATLAB R2014a.

[0083] The parameters in the simulation experiment are set as follows: the search window size is 7×7, the similarity window size is 3×3, the filter parameter h in NLM=1.0σ, the filter parameter h in HRW-NLM=4.0σ, h in HRWIMN-TSNLM (1) =4.0σ, h(2) =0.5σ(h (1...

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 provides an image denoising method based on mixed robust weights and method noise (MN), and belongs to the technical field of intelligent information processing. The method mainly solvesthe problem that balancing between noise suppression and detail retention is difficult to effectively maintain by a traditional NLM algorithm, and includes: firstly, using an improved mixed-robust-weight function to calculate similarity weights of image blocks; then using a pre-denoised image to construct the method noise, and carrying out combination with a two-level denoising framework; and finally, applying the provided mixed-robust-weight function and the method noise to a two-level non-local-mean (NLM) denoising method. The method of the invention is capable of effectively retaining structure detail information in the image while noise is suppressed, and has superior denoising performance.

Description

technical field [0001] The invention belongs to the technical field of intelligent information processing and relates to a non-local mean value denoising method in image denoising. Specifically, it is an image denoising method based on hybrid robust weight and method noise, which can be used in image denoising, computer vision, image analysis and other fields. Background technique [0002] Image denoising has always been a very basic and important research content in the field of image processing, which aims to effectively filter out the noise in the image and improve the visual effect of the image. At present, the more mature image denoising methods mainly include bilateral filtering, methods based on total variation, methods based on partial differential equations, methods based on wavelet threshold, etc. In 2005, Buades et al. proposed the Non-local Means (NLM) denoising algorithm for the first time, which has a wide range of applications in the fields of industry, agric...

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
CPCG06T5/70
Inventor 葛洪伟陆海青陈国俊
Owner JIANGNAN UNIV
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