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

Medical endoscopic image denoising method based on CBD-Net

An endoscopic image and medical technology, applied in the field of medical endoscopic image processing, can solve the problems of insignificant denoising effect, slow speed, inability to identify noise features, etc., and achieve the effect of improving denoising effect and generalization ability

Pending Publication Date: 2020-10-30
NANJING TUGE HEALTHCARE CO LTD
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Although bilateral filtering performs better in retaining edge parts, it cannot identify noise features, and cannot distinguish between high-frequency noise and high-frequency details. When the input noise parameter is too large, it is easy to erase the details, and vice versa. The noise effect is not obvious; and because bilateral filtering requires the gray level information of each center point field to determine its parameters, its speed is relatively slow, and the calculation amount grows at the square of the kernel size

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
  • Medical endoscopic image denoising method based on CBD-Net
  • Medical endoscopic image denoising method based on CBD-Net
  • Medical endoscopic image denoising method based on CBD-Net

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0061] Embodiments of the present invention will be described in further detail below in conjunction with the accompanying drawings.

[0062] According to the existing endoscopic image data, different algorithms are used to try to denoise to analyze its noise characteristics: Observing the existing endoscopic image, it can be seen that in practical applications, the ambient brightness of the endoscopic image is low, and the noise is more obvious, mainly Gaussian noise , so traditional methods such as Gaussian filtering, BM3D, and bilateral filtering, which perform well in Gaussian denoising, are used for preliminary denoising attempts. Denoising attempts with typical DnCnn, DnCnn-B, Ffd-Net, CBD-Net models.

[0063] Screen out the traditional and neural network methods with good performance for improvement and comparison: the traditional algorithm with good performance is BM3D, and the denoising network with good performance is FFDNet and CBD-Net.

[0064] Select CBD-Net acco...

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 medical endoscopic image denoising method based on CBD-Net, and the method comprises the steps: constructing and improving CBD-Net, and obtaining a medical endoscopic image denoising model; obtaining a noise source according to an imaging principle of the endoscope; obtaining a simulation training set and a real training set, and obtaining a supplementary training set according to a noise generation source; training a medical endoscopic image denoising model by adopting the simulation training set, the real training set and the supplementary training set; and denoising the medical endoscopic image by adopting the trained medical endoscopic image denoising model. According to the invention, the original hue, brightness and other factors of the endoscope image are not changed while the endoscope image is processed; moreover, the details of the medical image are well reserved.

Description

technical field [0001] The invention belongs to the technical field of medical endoscopic image processing, in particular to a method for denoising medical endoscopic images based on CBD-Net. Background technique [0002] Medical electronic endoscope is the product of the continuous development and fusion of traditional endoscope and computer, microelectronics and other technologies. The lesion is used as a sample to provide more information for the next step of treatment. However, in practical applications, medical images are often affected by imaging equipment and the external environment during the acquisition, conversion, and transmission. Endoscopic images will inevitably have some noise. The existence of noise will seriously reduce the image quality and damage the image quality. Detailed information interferes with the doctor's correct judgment and even leads to misdiagnosis. Therefore, how to improve the endoscope denoising technology so that it can remove the noise...

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): G06T7/00G06T5/00
CPCG06T7/0012G06T2207/10068G06T2207/20081G06T2207/20084G06T5/70G06T5/20G06T5/60
Inventor 汪彦刚张诚天陈阳
Owner NANJING TUGE HEALTHCARE CO LTD
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