CNN medical CT image denoising method based on noise prior

A CT image and noise technology, applied in the field of medical image denoising, can solve the problems of loss of detailed images, incompatible denoising, and difficult multi-level denoising of detailed textures, so as to achieve enhanced robustness, enhanced generalization ability, The effect of increased speed

Pending Publication Date: 2021-02-26
ZHEJIANG UNIV OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, some detailed images will be lost in the image after noise reduction, and more detailed textures will be lost in relatively high noise images. It is also difficult to achieve true multi-level denoising, which does not meet the requirements of denoising. Purpose

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
  • CNN medical CT image denoising method based on noise prior
  • CNN medical CT image denoising method based on noise prior
  • CNN medical CT image denoising method based on noise prior

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] Below in conjunction with accompanying drawing, the present invention is specifically explained

[0050] The concrete steps of CNN medical CT image denoising algorithm based on noise prior of the present invention are as follows:

[0051] Step 1) Create a medical CT image model:

[0052] Create a medical CT image model using a Gaussian noise model. Its mathematical expression is:

[0053] Y=X+V (1)

[0054] Among them, X is a clean image without noise, Y is the actual noise-containing image, and V is noise; the noise distribution of V obeys Gaussian distribution, and Gaussian noise refers to the probability density function of which obeys Gaussian distribution (that is, normal distribution). Noise-like, noisy images such as figure 1 As shown, it is the Gaussian random variable z probability density function, and its mathematical expression is:

[0055]

[0056] Among them, μ is expressed as mathematical expectation, and σ is expressed as standard deviation;

[...

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 CNN medical CT image denoising based on noise prior. The method comprises the following specific steps: 1), creating a medical CT image model; 2) constructing a noise prior information extraction network; 3) constructing a denoising network; 4) training a noise prior information extraction network and updating parameters; 5) training a training denoising network and updating parameters; 6) denoising the medical CT image; the medical CT image denoising method has the advantages and innovations that denoising is carried out by using noise prior information of the medicalCT image; two networks are used for denoising, wherein the two networks are respectively a noise prior information extraction network and a denoising network; performing the noise information extraction on the CT medical image by the noise prior information extraction network to obtain more image information; serially splicing a noise level diagram predicted by the noise prior information extraction network with the noise diagram, and inputting the noise level diagram and the noise diagram into a denoising network; the BN layer is added to the network, so that the generalization ability of the network is improved, the network convergence speed is increased, and the network denoising performance is improved.

Description

technical field [0001] The present invention relates to the field of medical image denoising, mainly relates to medical CT images, in particular to a noise prior-based CNN medical CT image denoising method suitable for medical CT images. [0002] technical background [0003] Because medical CT images can easily and conveniently obtain patient data information, medical image processing has gradually attracted people's attention. Medical CT images obtained by using various technical means have become an indispensable part of the work and life of medical staff, helping doctors to conduct accurate analysis and comprehensive diagnosis of diseases. With the continuous increase in the amount of patient data currently obtained, the symptoms are gradually showing a trend of diversity, which brings great challenges to patients in making scientific medical diagnosis and accurate treatment. [0004] During the process of digital data acquisition and data transmission, medical CT images...

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): G06T5/00G06N3/04G06N3/08
CPCG06N3/08G06T2207/10081G06T2207/20081G06T2207/20084G06T2210/41G06N3/045G06T5/70
Inventor 张聚牛彦施超潘玮栋陈德臣
Owner ZHEJIANG 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