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

Non-local mean denoising optimization method based on structural similarity

A non-local average and structural similarity technology, which is applied in image data processing, instrumentation, computing, etc., can solve problems such as large differences in denoising performance with different noise intensities, blurred image details, and decreased ability to denoise strong noise

Active Publication Date: 2017-10-03
WUHAN UNIV OF SCI & TECH
View PDF2 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The core problem of the non-local mean filtering algorithm is to determine the weighted kernel function. The original non-local mean denoising algorithm uses an exponential kernel function for weighting, which leads to excessive smoothing and blurring of image details, and the pure exponential or cosine kernel function It cannot adapt to the change of noise, and the denoising ability of strong noise is obviously reduced; the improved quadratic exponential kernel function makes the weighting value decrease rapidly with the increase of distance, resulting in poor processing ability of image details and different noise intensity The denoising performance varies greatly, etc.

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
  • Non-local mean denoising optimization method based on structural similarity
  • Non-local mean denoising optimization method based on structural similarity
  • Non-local mean denoising optimization method based on structural similarity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0064] A Structural Similarity Based Non-Local Mean Denoising Optimization Method. The concrete steps of described method are:

[0065] Step 1. Select a picture such as figure 1 The noise-contaminated image X shown, figure 1 is a pair of noise-contaminated images to be denoised in this embodiment. The size of the noise-contaminated image X is 256×256, and the noise standard deviation is 25. Let any pixel point i, j∈X, (m, n) be the coordinates of any pixel point, select a pixel point i in the noise-contaminated image X, and use the pixel point i as the center to establish a 7×7 noise image search box.

[0066] Step 2. Take a 3×3 noise image similarity frame X in the noise image search frame t , with the noisy image similar to box X t Swipe in the noise image search box to find out all noise image similar frames X in the noise image search box t The combination of each noise image similarity frame X is recorded t The pixel point j in the center is in the noise image sea...

Embodiment 2

[0119] A Structural Similarity Based Non-Local Mean Denoising Optimization Method. The concrete steps of described method are:

[0120] Step 1. Select a picture such as image 3 The noise-contaminated image X shown, image 3 is a pair of noise-contaminated images to be denoised in this embodiment. The size of the noise-contaminated image X is 256×256, and the noise standard deviation is 25. Let any pixel point i, j ∈ X, (m, n) be the coordinates of any pixel point, select a pixel point i in the noise-contaminated image X, and set up a 7×7 pixel point i as the center noise image search box.

[0121] Step 2. Take a 3×3 noise image similarity frame X in the noise image search frame t , with the noisy image similar to box X t Swipe in the noise image search box to find out all noise image similar frames X in the noise image search box t The combination of each noise image similarity frame X is recorded t The pixel point j in the center is in the noise image search box cent...

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 relates to a non-local mean denoising optimization method based on structural similarity. The technical scheme is characterized by obtaining a noise pollution image X, and calculating a weighted value w(i,j) of all pixel points j with respect to a pixel point i; carrying out primary denoising on the noise pollution image X through a weight kernel function improved non-local means method to obtain a primarily-denoised image X<^>; in the primarily-denoised image X<^>, calculating weighted value wSSIM(i,j) of all primarily-denoised image similar boxes X<^>j with respect to d*d primarily-denoised image similar boxes X<^>i with the pixel point i being as the center; and adjusting all of noise image similar boxes in the noise pollution image X, and carrying out secondary denoising on the noise pollution image X to obtain a secondarily-denoised image X<->. The method has the advantages of being capable of keeping image details, suitable for different noise intensity and good in visual quality.

Description

technical field [0001] The invention belongs to the technical field of non-local mean denoising optimization. Specifically, it relates to a non-local mean denoising optimization method based on structural similarity. Background technique [0002] Image denoising is the most basic and widely researched hot issue in the field of image processing, and its purpose is to remove various noise pollution introduced in the process of image acquisition and transmission. Among many image denoising algorithms, the original non-local mean algorithm proposed by Buades et al. has been proved to outperform other classical denoising methods. The basic idea of ​​this algorithm is to use a large amount of redundant information in the image to perform a global search on the image block where each pixel in the image is located to find similar blocks, and reduce random noise by weighted average of similar structural blocks to achieve denoising effect. [0003] The core problem of the non-local...

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
CPCG06T2207/10004G06T2207/20024G06T2207/20021G06T5/70
Inventor 柴利张璐盛玉霞
Owner WUHAN UNIV OF SCI & TECH
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