Feature loss-based medical image super-resolution reconstruction method

A technology for super-resolution reconstruction and medical image, applied in the field of medical image super-resolution reconstruction based on feature loss, which can solve problems such as checkerboard and lattice artifacts, and achieve a wide range of applications.

Inactive Publication Date: 2017-11-21
CHENGDU UNIV OF INFORMATION TECH
View PDF3 Cites 47 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is that during traditional super-resolution reconstruction, artifacts similar to checkerboard grids will be generated. The purpose is to provide a method for medical image super-resolution reconstruction based on feature loss, and to solve the problem of Issue with artifacts resembling a checkerboard

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
  • Feature loss-based medical image super-resolution reconstruction method
  • Feature loss-based medical image super-resolution reconstruction method
  • Feature loss-based medical image super-resolution reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] A Feature Loss-Based Method for Medical Image Super-Resolution Reconstruction, Including Image Transformation Network f w , the image conversion network f w It is a feedforward fully connected neural network. The feedforward neural network divides each neuron in the network into different groups according to the order of receiving information. Each group is regarded as a network layer, and the neurons in each layer receive The number output of the previous layer of neurons is used as its own input, and then its own output is input to the next layer, and the information in the entire network is propagated in one direction; the image conversion network f w Process a low-resolution medical image of size H / 4×W / 4, and convert the low-resolution medical image of size H / 4×W / 4 into a high-resolution image of size H×W, where H and W are natural numbers . The image conversion network f w Including two upscaling convolutional layers, after two upscaling convolutional layers, th...

Embodiment 2

[0032] A Feature Loss-Based Method for Medical Image Super-Resolution Reconstruction, Including Image Transformation Network f w , the image conversion network f w It is a feedforward fully connected neural network. The feedforward neural network divides each neuron in the network into different groups according to the order of receiving information. Each group is regarded as a network layer, and the neurons in each layer receive The number output of the previous layer of neurons is used as its own input, and then its own output is input to the next layer, and the information in the entire network is propagated in one direction; the image conversion network f w Process a low-resolution medical image of size H / 4×W / 4, and convert the low-resolution medical image of size H / 4×W / 4 into a high-resolution image of size H×W, where H and W are natural numbers . The image conversion network f w Including two upscaling convolutional layers, after two upscaling convolutional layers, th...

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 feature loss-based medical image super-resolution reconstruction method. The method involves an image conversion network fw; the image conversion network fw is a feed-forward full connection neural network; neurons in the feed-forward full connection neural network are divided into different groups according to an information receiving sequence; each group is regarded as a network layer; the neurons in each layer receive numerical outputs of the neurons in the previous layer to serve as inputs of themselves, and outputs of themselves are input to the next layer; information in the whole network is transmitted in a direction; and the image conversion network fw receives a low-resolution medical image having a size of H/4XW/4 and sent by the feed-forward neural network, and converts the low-resolution medical image having the size of H/4XW/4 into a high-resolution image having a size of HXW.

Description

technical field [0001] The invention relates to a super-resolution reconstruction, in particular to a medical image super-resolution reconstruction method based on feature loss. Background technique [0002] In recent years, there are mainly two types of super-resolution reconstruction methods: one is the reconstruction-based method; the other is the learning-based method. The reconstruction-based method is to model the acquisition process of the low-resolution image, use the regularization method to construct the prior constraints of the high-resolution image, and estimate the high-resolution image from the low-resolution image, which is lost in the process of reconstruction and degradation. Finally, the problem is transformed into the problem of cost function optimization under constraints; the other type is learning-based super-resolution reconstruction, the basic idea of ​​which is to obtain high-resolution images and ground-resolved images through learning. In this typ...

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): G06T11/00G06T3/40
CPCG06T3/4046G06T3/4053G06T11/008G06T2207/10081G06T2207/10088G06T2207/30016
Inventor 符颖吴锡邢晓羊李玉莲周激流
Owner CHENGDU UNIV OF INFORMATION 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