NCSR model-based three-dimensional collaborative filtering and denoising method

A collaborative filtering and model technology, applied in the field of image processing, can solve problems such as image edge ringing effect, and achieve the effect of reducing halo and ringing effect

Inactive Publication Date: 2018-06-22
XIAN UNIV OF TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When filtering non-local image blocks, there is often a ringing effect at the edge of the image

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
  • NCSR model-based three-dimensional collaborative filtering and denoising method
  • NCSR model-based three-dimensional collaborative filtering and denoising method
  • NCSR model-based three-dimensional collaborative filtering and denoising method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] The three-dimensional collaborative filtering denoising method based on the NCSR model of the present invention, such as figure 1 shown, follow the steps below:

[0043] Step 1, utilize the Canny operator to extract the edge pixels of the noisy image, and save the coordinates of the edge pixels in the array W;

[0044] The classification of pixels is divided into two categories, one is the pixels in the smooth area, and the other is the pixels in the edge area;

[0045] Step 2, grouping the noisy images in step 1 by block matching to obtain a plurality of grouped image blocks;

[0046] The specific steps of step 2 are: first determine the position of the reference block, starting from the first 8×8 image block in the upper left corner of the image as the first reference block, and the position of the next reference block is to translat...

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 NCSR model-based three-dimensional collaborative filtering and denoising method. The method specifically comprises the following steps of: 1, extracting edge pixel points ofimages with noise, and storing coordinates of the edge pixel points in W so as to realize pixel classification; 2, grouping the images with noise in the step 1 through a block matching manner; 3, carrying out collaborative filtering and denoising on each group of image blocks in the step 2; and 4, carrying out weighted average on the images denoised in the step 3 and carrying out aggregation to solve final estimated values of the images, and carrying out regional processing on the images with noise so as to further improve the clearness degrees of detail parts of the images. According to theNCSR model-based three-dimensional collaborative filtering and denoising method, noise is filtered by utilizing a filter formula in an NCSR model, so that non-local similarity of the images is sufficiently utilized, and the halo and ring effects in denoising results are obviously decreased.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a three-dimensional collaborative filtering denoising method based on an NCSR model. Background technique [0002] In recent years, many scholars and researchers have continuously proposed new denoising methods by studying the characteristics of image noise and its own signal. The common goal of these methods is to retain more image details while achieving denoising. For a natural image, there is a great correlation between its structure and texture features. Therefore, whether it is to use the spatial domain method to directly average the pixel values, or to use the frequency domain method to transform the image and then perform threshold processing on the transform coefficients, both local and non-local correlation characteristics of the image itself are fully utilized to obtain relatively Good denoising effect. [0003] The methods of denoising by using local simil...

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
CPCG06T5/70
Inventor 刘晶刘睿娇陈进磊
Owner XIAN UNIV OF 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