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

An Electromagnetic Inverse Scatter Imaging Method Based on Perceptual Generative Adversarial Network

An imaging method and inverse scattering technology, applied in biological neural network models, neural learning methods, radio wave reflection/re-radiation, etc. The effect of improving image quality

Active Publication Date: 2022-03-22
HEFEI UNIV OF TECH
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If there is no explicit constraint on the features of the target image, it is easy to cause artifacts in the reconstructed 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
  • An Electromagnetic Inverse Scatter Imaging Method Based on Perceptual Generative Adversarial Network
  • An Electromagnetic Inverse Scatter Imaging Method Based on Perceptual Generative Adversarial Network
  • An Electromagnetic Inverse Scatter Imaging Method Based on Perceptual Generative Adversarial Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] In this embodiment, an electromagnetic inverse scatter imaging method based on perceptual generative adversarial networks, first uses the backpropagation method to generate a low-resolution scatterer image x from the measured scattering field, and then uses the generator G θ Mapping from low-resolution images to target images to generate near-real reconstructed images G θ (x). The low-resolution image x generated by BP is used as a condition of the discriminator and the reconstructed image G θ (x) or the real target image y is paired into the discriminator, where the discriminator is used as a feature extractor to extract the features of different hidden layers, so that the reconstructed image and the target image are matched in terms of pixels and features at the same time.

[0040] Specifically, if figure 1 shown, including the following steps:

[0041] Step 1. Data generation;

[0042] Step 1.1. In a two-dimensional transverse magnetic field, assuming the frequen...

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 electromagnetic inverse scattering imaging method based on perception-generated confrontation network, the steps of which include: 1. quickly generating a low-resolution scatterer image by using the backpropagation method according to the measured scattering field; 2. network structure construction stage , using the architecture of generative adversarial network, design the structure of generator and discriminator; 3 loss function design stage, add perceptual adversarial loss in the objective function, use the hidden layer of the discriminator to extract perceptual adversarial loss, so that the reconstructed image and the target image are simultaneously in pixel Matching with features; 4. Reconstruction of relative permittivity of scatterers by training perceptual generative confrontation network. The invention can more effectively enable the generation network to learn target feature information, thereby better improving the imaging quality.

Description

technical field [0001] The invention belongs to the technical field of electromagnetic inverse scattering imaging, and in particular relates to an electromagnetic inverse scattering imaging using a deep learning method. Background technique [0002] Electromagnetic inverse scattering uses the measured scattering field combined with the inversion algorithm to determine the location, shape and physical parameters of the scatterer. Usually electromagnetic inverse scattering is a highly nonlinear and ill-posed problem. After years of development, researchers have proposed various electromagnetic inverse scattering reconstruction algorithms, among which the quantitative method is the mainstream direction of electromagnetic inverse scattering research because it can obtain all the information of the scatterers. [0003] Quantitative methods generally define a nonlinear objective function including regularization terms, and then use global or local linearization approximation to i...

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 Patents(China)
IPC IPC(8): G01S13/89G01S7/41G06N3/04G06N3/08
CPCG01S13/89G01S7/418G01S7/417G06N3/084G06N3/048G06N3/045
Inventor 宋仁成黄优优刘羽李畅成娟陈勋
Owner HEFEI UNIV OF 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