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

Three-dimensional magnetic layer reconstruction method and system based on limited angle projection data

A projection data, limited angle technology, applied in the field of spatial detection projection reconstruction, can solve the problem of loss of detail information, distortion, etc., to achieve the effect of strengthening generalization ability and improving accuracy

Active Publication Date: 2022-04-05
NAT SPACE SCI CENT CAS
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Figure 8(a) is a cross-sectional view of the full-angle projection data, Figure 8(b) is a cross-sectional view of the limited-angle projection data, Figure 8(c) is the FDK reconstruction result of the full-angle projection data, and Figure 8(d) is the limited-angle projection data FDK reconstruction results; From the results point of view, the reconstruction results using limited-angle projection data have been severely distorted, and detailed information has also been seriously lost

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
  • Three-dimensional magnetic layer reconstruction method and system based on limited angle projection data
  • Three-dimensional magnetic layer reconstruction method and system based on limited angle projection data
  • Three-dimensional magnetic layer reconstruction method and system based on limited angle projection data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0083] The full name of the generative confrontation network is Generative Adversarial Networks, referred to as GAN. It is an unsupervised learning method proposed by Goodfellow et al. in 2014. It includes two parts: the generator network and the discriminator network. The generator network is dedicated to generating. Realistic data comparable to real data, the discriminator is dedicated to discriminating between real data and generator-generated data. The two networks play games with each other. The generator is forced to generate seemingly real samples out of continuous learning of real data under pressure, and the discriminator further learns to distinguish generated samples from real samples. In 2016, AlecRadford proposed Deep Convolution Generative Adversarial Networks (DCGAN) on the basis of GAN. The network structure diagram of DCGAN is similar to that of GAN, but both the generation network and the discriminant network use depth. Convolutional neural network, and remov...

Embodiment 2

[0143] Embodiment 2 of the present invention proposes a three-dimensional magnetosphere reconstruction system based on limited-angle projection data, which is implemented based on the method in embodiment 1. The system includes: a missing-angle three-dimensional projection data generation module, a completion module and a three-dimensional reconstruction module ;in,

[0144] The missing-angle three-dimensional projection data generating module is used to receive three-dimensional magnetospheric X-ray projection data of a limited angle, and input the pre-established and trained 3D generation confrontation network to obtain three-dimensional projection data of missing angles;

[0145] The completion module is used to optimize the three-dimensional projection data generated by the 3D generation confrontation network based on the projection optimization completion network, so as to realize the data completion of the missing angle of the three-dimensional projection data;

[0146] ...

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 three-dimensional magnetic layer reconstruction method and system based on limited angle projection data. The method comprises the following steps: receiving three-dimensional magnetic layer X-radiation projection data of a limited angle, and inputting the data into a pre-established and trained 3D generative adversarial network to obtain three-dimensional projection data of a missing angle; optimization is carried out from the three-dimensional projection data generated by the 3D generative adversarial network based on a projection optimization and completion network, and data completion of the missing angle of the three-dimensional projection data is realized; and realizing three-dimensional reconstruction based on the complemented three-dimensional projection data. Compared with the prior art in which single angle integral projection data is utilized, the method has the advantages that three-dimensional CT reconstruction is carried out on the magnetic layer by utilizing multi-angle integral projection information; the research blank of three-dimensional magnetic layer computed tomography (CTA) is filled, and the accuracy of three-dimensional reconstruction is improved; the evaluation function of the projection completion optimization network is completely innovated, and the mean square error is minimized while the structural similarity is kept.

Description

technical field [0001] The invention relates to the technical field of space detection projection reconstruction, in particular to a three-dimensional magnetosphere reconstruction method and system based on limited-angle projection data. Background technique [0002] The Solar Wind-Magnetosphere Interaction Panoramic Imaging Satellite Project (SMILE) will use innovative X-ray and ultraviolet imaging instruments to provide the first global image of the interaction between the solar wind and the Earth's magnetosphere. Such as figure 1 A schematic diagram of the SMILE satellite mission is shown. [0003] However, since the SMILE satellite has not yet been launched, and there is no magnetospheric X-ray imaging satellite launched yet, there is no X-ray observation data of the magnetosphere. But solar physicists have created the PPMLR-MHD model, which is used to simulate the solar wind-magnetosphere-ionosphere coupling system, and can use this model to conduct preliminary studie...

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): G06T17/00G06N3/08G06N3/04G06K9/62G06F30/27G06V10/74G06V10/774G06V10/82
Inventor 王荣聪李大林彭晓东孙天然
Owner NAT SPACE SCI CENT CAS
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